1,577 92 1MB
Pages 254 Page size 432 x 648 pts Year 2004
Learning in Science
Learning in Science brings together accounts of the five influential and groundbreaking Learning in Science Projects undertaken by the author and colleagues over a period of twenty years. Offering comprehensive coverage of the findings and implications of the projects, the book offers insight and inspiration at all levels of science teaching and learning, from primary and secondary school science to teacher development and issues of classroom assessment. The book reviews the findings in the light of current science education debates, and is thematically organised to illuminate continuous and emerging themes and trends, including: • • • • •
learning pedagogy assessment Maori and science education curriculum development as teacher development.
Learning in Science will be a valuable resource for science teachers, science teacher educators, science education researchers, curriculum developers and policy makers. Beverley Bell is Associate Professor in the School of Education at the University of Waikato with research and teaching interests in pedagogy, learning, assessment and teacher education.
Learning in Science
The Waikato Research
Beverley Bell
First published 2005 by RoutledgeFalmer 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Simultaneously published in the USA and Canada by RoutledgeFalmer 270 Madison Ave, New York, NY 10016 This edition published in the Taylor & Francis e-Library, 2004. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” RoutledgeFalmer is an imprint of Taylor & Francis © 2005 Beverley Bell All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0-203-46499-0 Master e-book ISBN
ISBN 0-203-47318-3 (Adobe eReader Format) ISBN 0-415-29874-1 (hbk) ISBN 0-415-29875-X (pbk)
This book is dedicated to Martin and Laura who gave me a purpose for doing the research and development and in memory of my parents, Frank and Kathleen who valued education and supported my studies and Nick who understood horses, trucks and Holdens
Contents
List of illustrations Preface Acknowledgements
viii ix xii
1 An introduction and overview
1
2 Constructivist views of learning
17
3 Social views of learning
40
4 Pedagogies that take into account students’ thinking
61
5 More pedagogies
90
6 Classroom assessment of science learning
116
7 Culture, Maori and science education: the social, cultural, political and historical contexts
144
8 Culture, Maori and science education: research and development
164
9 Science curriculum development and teacher development
180
Glossary of MAori words References Index
202 205 235
List of illustrations
Figures 2.1 The everyday and the scientific meaning of animal 4.1 The Generative Model of Teaching 4.2 Children’s models of electric current 4.3 The Interactive Teaching Approach 6.1 What is assessed in science classrooms? 6.2 Planned formative assessment 6.3 Interactive formative assessment 6.4 A model of formative assessment 9.1 A model of teacher development 9.2 Teacher development and curriculum development
19 71 72 75 120 125 125 126 183 196
Table 6.1 Planned and interactive formative assessment
127
Preface
The Waikato river is the longest and most developed river in Aotearoa, New Zealand, running through the land of Maori iwi of Tuwharetoa, Te Arawa and Tainui. The journey of 425 km begins on the slopes of Mt Ruapehu and the Tongariro National Park, on the central north island plateau, feeds into Lake Taupo, threads its way through the forestry plantations of Pinus radiata, and then through the Waikato basin and the city of Hamilton, past Taupiri, a sacred mountain for Maori, to the Waikato delta and Port Waikato, where it runs into the Tasman Sea, on the west coast just south of Auckland. The river is joined by numerous tributaries along its course, its main one being the Waipa river. The river sustains nine hydro-power stations and two geothermal stations. It has a gentle gradient but turbulent falls (such as the famous Huka falls) and wide sweeping stretches (such as around the Meremere district) which travellers see on their way from Auckland International Airport to Hamilton city. It is a river that brings life to the people who live along the river in Putararu, Cambridge, Hamilton, to those farming the area in forestry, dairy, cattle, race horses and deer, to the trout hatcheries and fishing, to the numerous bird varieties, and to the gardens and orchards of the region. Maori have lived along the Waikato and Waipa rivers since around 1500. The Tainui waka arrived from East Polynesia to the west coast of the north island at Mokau, then Kawhia, between 1100 and 1300 (King, 2003). Later Tainui moved inland to the Waikato and Waipa rivers and wetlands which became their main food sources. In the 1840s and 1850s (before the Land Wars and European settlement in the 1860s) Tainui had extensive gardens along the Waikato of kumera, yam, taro, and hue (gourd). By the 1850s, 2000 waka (canoes) were transporting produce to Auckland markets via the Waikato river. The river is sacred to Waikato Maori: Ngati Tuwharetoa and Tainui (the Tainui Federation which has four main divisions – Pare Hauraki, Ngaati Raukawa, Ngaati Maniapoto and Waikato) who see the river and the people as one. There is no separation. The river is enmeshed in their history, stories and world view, and the waters of the river were important for spiritual and physical cleansing. The Waikato river is a kaitiaki (guardian) to Tainui people. It is seen as a living entity; the river is alive. A healthy river means a healthy people – spiritually as well as physically.
x
Preface
My family arrived on the banks of the Waikato river in 1869. My paternal great-great-grandfather, John Bell, emigrated as a boy from Gillingham, Dorset, England, arriving in New Zealand in 1842, two years after the founding document of New Zealand, the Treaty of Waitangi, was signed. He settled with his wife Mary Anne and 13 surviving children in Cambridge, just south of Hamilton, working as a sawyer and baker. Seven born-generations of my family have now lived on the land surrounding the Waikato river. My family’s journey has been one of prosperity, with the family farm of 34 years in the Puketaha district, on the outskirts of Hamilton [part of the orginal Woodlands estate (King, 2003) ] having been one of the largest town milk supply dairy farms in New Zealand in the 1950s–60s (Williamson, 1954). But this prosperity was not shared by Maori. My great-great-grandfather was part of the first wave of English settlers, who came to farm the fertile Waikato land in the late 1860s. Much of the Tainui land was illegally confiscated by the government or bought in dubious deals from Tainui Maori after the Land Wars (King, 2003), before being sold on to the settlers such as my family. As a consequence, Tainui Maori lost land and livehoods, and over the years language and cultural practices as well. In compensation for the land, the Crown returned crown-owned land to Tainui in 1996, including the land on which the University of Waikato now sits (McCan, 2001). Science education research at the University of Waikato has also had a long journey. It began when Roger Osborne, a former high school teacher of senior physics and foundation lecturer in the Physics Department of the then newly established University of Waikato, did his doctoral thesis under the supervision of Peter Freyberg, in the 1970s. It has continued over 25 years with five Learning in Science Projects, doctoral and masters theses and related staff research. It has energised and powered the careers of many New Zealand science educators. Many overseas colleagues have visited not once, but several times. It has had turbulent times such as the sudden deaths of Roger Osborne and Peter Freyberg, in 1985, and times of celebration as when the first Learning in Science Project team was awarded the Royal Society Science and Technology Medal in 1994. It has had calm times when staff and students quietly got on with the more routine aspects of doing research. Science education research was first formally recognised by the university on the establishment of the Science Education Research Unit in 1980, and despite three name changes for the Unit/Centre, science education has always been steadfastly there. The Waikato river and science education at the University of Waikato mean different things to different people. On a spring day in the early 1980s, a group of us floated down the Waikato river. I was in my blue kayak – a cruising one with a small brass keel for ease of paddling straight or not paddling at all. Roger Osborne, Ross Tasker, John Happs and Keith Stead were in an assortment of rubber dingies, inner tubes of car tyres and kayaks. We drifted downstream from the Narrows, a narrow kink in the river, just south of Hamilton, to a beach in a northern suburb of the city. The journey took us several hours, during
Preface
xi
which time there was much laughter and chatting. Each of us would have noticed and remembered different things of the journey. As a biology teacher, I noted the yellow flowers of the native kowhai and the thick stands of more recent arrivals such as the willows. No doubt John noted the rock formations in the cliffs. Each of us would be able to recall the same and different bits of the day. Likewise, each of us will have different stories to tell about the accumulated research of science education at the University of Waikato, as well as the stories to tell about the Waikato research in technology education and environmental education. But these stories are for other books. This book tells one version of the past twenty years (1979–98) of science education research at the University of Waikato – the story of one person’s journey, a journey made with many. And for the many who continue the work in science education at Waikato. This book will hopefully serve as a collective reference for the ongoing research, as a background to current debates in science education in New Zealand and internationally, and as a textbook for students. New Zealand is a bicultural nation, having two official languages, English and Maori. Hence, either English or Maori can be used in official communications in, for example, schools, parliament or the courts of law. Accordingly, both Maori and English words are used in this book, with a translation of the Maori into English, in brackets, as need be. In this way, both Maori and English are given equal status as first languages in the book and Maori is not positioned as ‘other’. A glossary of Maori terms used is given at the end of the book. Beverley Bell
Acknowledgements
I wish to acknowledge and thank the following people who have contributed to my Waikato journey in science education over the last twenty years: The students and teachers of the Waikato and New Zealand schools, whose ongoing support and involvement were central to the Waikato research in science education programme. Colleagues and students with whom I have worked: Robyn Baker, Miles Barker, Jane Barnett, Bill Barton, Caroline Bennett, Di Bentley, Fred Biddulph, Russell Bishop, Paul Black, Sally Brown, Malcolm Carr, Margaret Carr, Wheijen Chang, May Cheng, Meiying Chu, Guy Claxton, Bev Cooper, Mark Cosgrove, Bronwen Cowie, Terry Crooks, Christine Ditchfield, Wendy Drewery, Rosalind Driver, Chris Eames, Gaalen Erickson, Lorraine Evening, Bev Farmer, Peter Fensham, Teresa Fernandez, Ian Francis, Peter Freyberg, Jane Gilbert, John Gilbert, Wayne Gribble, Richard Gunstone, Mavis Haigh, Hassan Hameed, John Happs, Wynne Harlen, Eleanor Hawe, Derek Hodson, Mary Hill, Anne Hume, Alister Jones, Leenana Kent, Valda Kirkwood, John Leach, John Loughran, Rosalie McGowan, Liz McKinley, Pauline McPherson Waiti, Cathy Manthorpe, Sue Middleton, Abdul Muhsin Mohamed, Sinapi Moli, Jeff Northfield, Monti Ohia, Jonathon Osborne, Roger Osborne, Raewyn Oulton, Lynne Oxenham, John Pearson, David Porteous, Hazel Reddington, Leonie Rennie, Toby Rikihana, Susan Rodrigues, Sheema Saeed, Brendan Schollum, Phil Scott, Armajit Singh, Keith Stead, David Symington, Ross Tasker, Ian Taylor, Ken Tobin, Ian Torrie, James Ume, Timote Vaioleti, Tony Trinick, Mike Watts, Linda Wilkinson, Richard White, Zorina Wigglesworth, Rachel Wood. The Department and Ministry of Education for funding the ongoing research programme of the Learning in Science Projects and the NZ Council for Education Research, the British Council and the University of Waikato for supplementary funding. The management of the University of Waikato, especially Wilf Malcolm, Brian Gould, Charmaine Poutney, Noeline Alcorn, Roy Daniel, Ken Mackay, Margaret Nichols and the Study Leave Committee for the support, encouragement and opportunities to undertake and write up the research. Ngaere Roberts for her help with Te Reo Maori (the Maori language).
Acknowledgements
xiii
The administration, library and computer support people in the School of Education at the University of Waikato and in particular Jill Skerman, Steve Leichtweis, Yvonne Milbank, Aleisha Watson, Bernice Ziarno and John Wells. Acknowledgement is given to previous publications of the science education research at Waikato, including: Bell, B. F. (1993) Children’s Science, Constructivism and Learning in Science. Geelong, Victoria: Deakin University Press. Bell, B. and Baker, R. (eds) (1997) Developing the Science Curriculum in Aotearoa New Zealand. Auckland: Addison Wesley Longman (Pearson Education New Zealand). Bell, B. and Cowie, B. (2001) Formative Assessment in Science Education. Dordrecht: Kluwer Academic Press. Bell, B. and Gilbert, J. (1996) Teacher Development: A Model from Science Education. London: Falmer Press. Osborne, R. and Freyberg, P. (1985) Learning in Science: The Implications of Children’s Science. Auckland: Heinemann. Acknowledgement is also given for the use of material, with permission: Figure 4.1 and 4.2: In Osborne, R. and Freyberg, P. (1985) Learning in Science: The Implications of Children’s Science. Auckland: Heinemann (Reed Publishing). Figure 4.3: In Faire, J. and Cosgrove, M. (1988) Teaching Primary Science. Hamilton: Waikato Education Centre. Figures 6.1, 6.2, 6.3 and 6.4 and Table 6.1: In Cowie, B. and Bell, B. (1999) A model of formative assessment in science education, Assessment in Education, 6(1): 101–16. Figure 9.1: In Bell, B. (ed.) (1993) I Know about LISP but How do I Put it into Practice? Final Report of the learning in Science Project ( Teacher Development). Hamilton: University of Waikato and the Ministry of Education. Figure 9.2: In Bell, B. and Baker, R. (1997) Developing the Curriculum in Aotearoa New Zealand. Auckland: Addison Wesley Longman (Pearson Education New Zealand). I wish to thank the editorial staff at RoutledgeFalmer for their support and help: Anna Clarkson, Hywel Evans, and Jessica Simmons. I thank Derek for his encouragement during the writing of this book.
Chapter 1
An introduction and overview
1
An introduction and overview
The science education research programmes at the University of Waikato have spanned over twenty-five years, making a unique contribution to international science education research. This book reviews the research done in the first twenty years, between 1979 and 1998. The core of the research programmes has been not one, as is often commonly thought, but five projects, called the Learning in Science Projects, funded by the New Zealand Department/Ministry of Education, and associated masters and doctoral research. The five Learning in Science Projects (LISP) were:
LISP (Forms 1–4) LISP (Primary) LISP (Energy) LISP (Teacher development) LISP (Assessment)
Dates 1979–1981 1982–1984 1985–1988 1990–1993
Directors Osborne, Freyberg Osborne, Freyberg Carr Bell
Students’ age 11–14 years 7–10 years 11–17 years 11–14 years
1995–1996
Bell
11–14 years
A brief introduction to each project and related theses follows.
Learning in Science (F1– 4) Project (1979 – 81) Introduction This project was funded by the Department of Education, to research science education in Forms 1–4 (Years 7 to 10). The project began shortly after the 1978 Draft Science Syllabus and Guide for students in Years 7 to 10 (Department of Education, 1978) was distributed to New Zealand schools. The Department was keen to obtain research information on ‘what to teach when’ in response to the prevailing views of learning at the time, notably learning hierarchies (Gagné, 1970; Raven, 1967–68; White, 1974) and the Piagetian notion of developmental stages (Kubli, 1979; Shayer and Adey, 1981). Hence, the Learning in Science Projects began as research to inform national science curriculum policy and development.
2
An introduction and overview
Most of the findings of the first LISP project were written up in the wellknown book Learning in Science: The Implications of Children’s Science (Osborne and Freyberg, 1985) and related theses (Barker, 1986; Bell, 1984a; Cosgrove, 1989; Happs, 1984; Schollum, 1986; Stead (now Bell), 1980a; Stead, 1984). There were three main phases to the research (Osborne et al., 1981) – description, analysis and action. The first phase (1979): the exploratory phase: perceived problems and difficulties with science education in New Zealand In the exploratory phase, the researchers interviewed those involved in science education, particularly teachers. The findings of the exploratory phase were documented in a series of 13 working papers, which focused on problems and difficulties associated with experiments, knowledge, topics, the teacher, the learner, process skills, the syllabus, attitudes, classrooms, written resources, and equipment (Freyberg et al., 1979). The findings were summarised in the Final Report for the project (Freyberg and Osborne, 1982). The second phase (1980): looking at the problems The second phase of the project studied some of these problems in depth, namely. The existing knowledge students bring to the classroom – children’s science A major aspect of the second phase of the project was the (then) newly identified aspect of pedagogy to be considered – children’s science (Osborne and Freyberg, 1985). A total of 13 working papers (Osborne, Freyberg and Tasker, 1980) and subsequent theses and published articles documented the alternative (to the accepted scientific) conceptions held by students. The recognition that students brought these alternative conceptions to their learning and that these ideas were strongly held on to led to theorising about what pedagogical practices might be required if students’ thinking were to be taken into account (Gilbert, Osborne and Fensham, 1982; Osborne, 1982a; Osborne, Bell and Gilbert, 1983). The main focus of this theorising was that students’ existing ideas needed to be addressed and that these ideas interacted with the taught curriculum possibly to lead to unexpected outcomes. This theorising underpinned the action research in the third phase of the first LISP project. These findings, and their significance, are detailed in Chapter 2. A view of mind – personal constructivist view of learning Another key outcome of the second phase of the first LISP project was the development of a personal constructivist view of mind (Osborne and Wittrock,
An introduction and overview
3
1983, 1985). The importance of ‘context’ in understanding learning was also raised by the development of a constructivist view of learning and mind (Bell, 1993a). This research outcome is discussed in detail in Chapter 2. Children’s classroom experiences The second phase of the research also explored in depth the problems perceived with the ‘practical’ work in science lessons (Schollum, 1986; Tasker, 1980, 1982; Tasker and Osborne, 1983, 1985). There were two kinds of portrayals of children’s classroom experiences. One was the examination of a few lessons in considerable depth (Osborne and Tasker, 1980; Osborne, Tasker and Stead, 1979; Tasker and Osborne, 1982a, b, c). The second kind of portrayal of children’s classroom experiences was a much more detailed analysis of some forty lessons (Tasker, 1981). The main finding was the differing perceptions that teachers and students have of the same classroom experiences. The differences were in terms of the scientific context of the activity, the scientific purpose of the activity, the scientific design of the investigatory activity, doing the activity, getting the results, thinking about what was done and what happened, the impact of the experience on children’s views and the relationship to predetermined outcomes (Tasker and Freyberg, 1985). These findings are addressed in Chapter 5. Outlooks on science This strand of the research investigated students’ outlooks on science – their habits and attitudes to science. In particular, girls’ and Maori students’ outlooks on science were explored (Stead, 1982, 1983, 1984). These findings are detailed in Chapter 7. The third phase (1981): toward action research The third and final phase of the research used an action-research model of research to develop and research new pedagogies that took into account students’ thinking and a constructivist view of mind (Osborne, Freyberg and Tasker, 1982). This included involving teachers in the development of viable solutions to the identified problems. The action research covered four areas: • • • •
biology (Bell, 1981a); physics (Osborne and Schollum, 1981); chemistry (Osborne, Schollum and Russell, 1982); and classroom activities (Lambert and Tasker, 1982).
The documentation of the action-research findings included extensive guide material on pedagogical practice and new types of science learning activities for students, for example card games in the biology resources. The researching
4
An introduction and overview
of new pedagogical practices was also documented in theses: pedagogies for photosynthesis (Barker, 1985a, b, 1986); earth sciences (Happs, 1984); animal, plant, living (Stead (now Bell), 1980a), and classroom activities (Schollum, 1986); and in parallel research projects on mixtures (Cosgrove, 1982a, b); and electric current (Cosgrove and Osborne, 1985a, b; Cosgrove, Osborne and Forret, 1989; Osborne, 1983). The findings of the third phase are reviewed in Chapter 4.
The Learning in Science Project (Primary) 1982 – 84 Introduction The primary project was the second of the Learning in Science Projects, beginning straight after the Forms 1–4 project in 1982 and again funded by the New Zealand Department of Education. It arose out of a concern that the problems and difficulties faced by middle school students of science (Years 7 and 8) might be fruitfully addressed earlier – that is, in the primary school. There were also concerns about the need for professional development for teachers wanting to use the research findings in their teaching. The LISP (Primary) project initially had two main purposes (Osborne, 1982b): • •
to investigate the problems and difficulties that primary school children have in learning science, and to find ways of overcoming such difficulties; the ‘training’ of primary school teachers to teach science.
The process–skills approach to primary science was not taken as a focus as it was thought that the process–skills objectives of the then New Zealand primary science syllabus (Department of Education, 1979) reflected ‘the philosophical disarray of the listed process skills, the artificiality of isolating individual skills and the minimal interest of most teachers of science in giving prime emphasis to process skills in their teaching’ (Osborne, 1982b: 1). Instead, the focus was on the ‘learner, on the learner’s prior experiences, memories and knowledge, and on the influence of teaching on these constructions of the world’ (Osborne, 1982b: 3). Teachers and pedagogy However, the main focus developed into the teacher, teaching and pedagogy. The pedagogical aspects of science education were seen as arising from the research on the learner. The importance of the teacher and teaching was soon evident in the discussions and foci of the research as indicated in the final report (Osborne and Biddulph, 1985b). There was also a commitment to involving teachers ‘in clarifying problems and seeking solutions to these problems at the earliest possible stage’ (Osborne
An introduction and overview
5
and Biddulph, 1985b: 5). This indicated a valuing of the existing pedagogical knowledge and practice of teachers in both the research and learning processes. The development of solutions to the problems and difficulties of learning primary science were seen as requiring the input of pedagogical knowledge and expertise of teachers. The pedagogical aspects of primary science were also to be focused on in a consideration of the processes of teacher education and teacher development. The early findings of the project relating to the status of teaching and learning primary science in New Zealand were centred around the dilemmas facing teachers of primary science. ‘As a consequence of the exploratory phase of the project, it was our view that the key problem of teaching and learning science in the primary school was centrally related to teacher perceptions and teacher confidence’ (Osborne and Biddulph, 1985b: 14). The pivotal place of the teacher and her or his pedagogical knowledges, outlook and practices in improving the learning outcomes in primary science was recognised. The pedagogical knowledges and practices researched by the team were obviously determined by the project team’s aims and purposes for primary science education. The following position was taken: Our interpretation of the broad aim of the (then) primary school science syllabus is that children’s interest in the environment should be fostered and that children should be given the opportunity to raise questions and undertake investigations to find answers to these questions. (Osborne and Biddulph, 1985b: 17) This aim was elaborated during the project (Osborne et al., 1982; Symington, Osborne, Freyberg and White, 1982) and was interpreted to mean that it follows from this perspective that science is being taught in primary schools whenever teachers help and encourage children to: i)
take an interest in the world around them and take responsibility for their own learning, ii) ask useful and productive questions about that world, and iii) make better sense in their own minds as to how and why things behave as they do . . . iv) gain new experiences and be interested in a wider view of the world, and v) be interested in other people’s explanations of the world and how such explanations have been obtained. (Osborne and Biddulph, 1985b: 18–19) These aims determined the pedagogies developed during the project – pedagogies that differed from those determined by the aims and philosophies of discovery,
6
An introduction and overview
transmission and process views of teaching and learning primary science (Osborne and Biddulph, 1985b). Children’s questions, ideas and investigations The interrelatedness of theorising on learning, curriculum and pedagogy was elegantly shown in the sub-studies on the questions and associated ideas that children had in specific areas of content. The children’s questions, explanations and ways of investigating simultaneously gave the researchers (and teachers) an insight into the existing ideas and thinking that primary pupils of science were bringing to the lesson, a pedagogical strategy for teaching the content, and a theoretical framework for choosing the content. The content areas chosen were not the big conceptual ideas of the 1960s and 1970s science curricula, such as ‘energy’ (Bruner, 1977) but topics (Harlen and Osborne, 1983) that enabled children to generate ideas which: • • •
have significance for making sense of everyday events and can be paced in a meaningful social context; can be generated by many primary children and related to their own prior knowledge and experiences in a meaningful way; can be tested by children through simple investigations (including referring to books and experts about the findings of others) and will, therefore, help rather than hinder further learning in science. (Osborne and Biddulph, 1985b: 25)
These topics were contexts that enabled children to link their existing ideas with new scientific ideas. The project team researched children’s ideas and questions about: • • • • • •
rocks (Symington, Biddulph, Happs and Osborne, 1983); spiders (Biddulph, 1983a; Hawe, 1983); metals (Biddulph and Osborne, 1983); floating and sinking (Biddulph, 1983b); flowering plants (Biddulph, 1984); hot and cold (Appleton, 1984).
Hence, the researching and documentation of the children’s ideas and explanations (Osborne, 1985) were primarily for the pedagogical purpose of informing teachers’ pedagogical knowledge of students, and for the purpose of structuring the classroom curriculum. Exploring alternative teaching strategies The work on a perspective of primary science and the initial investigations of children’s questions led to the ‘development of a lesson framework, or teaching model, designed to help teachers use a more interactive teaching approach in
An introduction and overview
7
their classrooms’ (Osborne and Biddulph, 1985b: 34). The rationale for the model (Biddulph, Osborne and Freyberg, 1983a) is that this teaching approach is one based on a constructivist view of learning as the teaching approach takes into account students’ thinking, and in particular their questions and explanations. The teaching approach is also based on a humanist view of learning as children were being encouraged to take ownership of their learning – to have a sense of control over it and feel that the learning was making sense to them (Biddulph, 1989). The key parts are: preparation; exploratory activities; children’s questions; children’s investigations; and reflection (Biddulph, 1990b; Biddulph and Osborne, 1984) and are explored in greater detail in Chapter 4. The findings of the LISP (Primary) had limited impact on the official primary science curriculum (Department of Education, 1979). A video was made available for the professional development of teachers of science (Department of Education, 1988a). Related research can be found in two theses (Biddulph, 1989; Fernandez, 1991). The findings of the LISP (Primary) are found mainly in Chapter 4. An ongoing concern of the research team was the place of science in the Interactive Teaching Approach as well as in the primary curriculum (Osborne, 1984). For some teachers using the interactive approach, the ‘science’ became lost in the sea of language activities in the context of a topic such as spiders, rocks, metals. The increased teacher confidence in teaching science using this interactive approach may have been due to their not having to know and use scientific ideas, but relying on other resources, such as books or visiting ‘experts’. This debate on the role of the teacher’s pedagogical content knowledge is ongoing. For example, to what extent do teachers need to know the scientific ideas about rocks to help a child investigate their question on ‘rocks’? How can teachers assess the science learning outcomes of the students if they do not have the scientific ideas themselves? How do children’s questions structure the classroom curriculum in relation to the official curriculum? These questions were addressed in later LISP projects. These concerns about the role of the content of science led to the funding of a third Learning in Science Project on the teaching and learning of ‘energy’.
Learning in Science Project (Energy) 1985 – 88 Introduction The Learning in Science Project (Energy) was a three-year project spread over the years 1985–88 and funded by the New Zealand Department of Education (Kirkwood and Carr, 1988a). The brief for this third project was to investigate the teaching and learning of the concept of ‘energy’ to 5 to 18-year-olds (Kirkwood and Carr, 1989). Whereas the first LISP project had investigated teaching and learning science by 11 to 14-year-olds, and the second project explored science for 7 to 10-year-olds, the third project investigated the teaching and learning of
8
An introduction and overview
a particular science concept across the broad age range of primary and secondary schooling. One concept that spanned the official primary and secondary science curricula was that of ‘energy’ (Department of Education, 1978, 1979). It was also hoped that this project would enable the Interactive Teaching model to be used and researched at the secondary level (Kirkwood et al., 1985) and for the science content to be given a focus during the researching of the teaching. There was one related thesis (Kirkwood, 1988). Research design The project had three phases: •
•
An initial phase explored the national and international literature; analysed the science syllabi, and included interviews of science educators (Kirkwood et al., 1985; Welch, 1985); Phase 2 was undertaken in junior secondary science classrooms and investigated the teaching and learning of ‘energy’ in actual classrooms. Three in-depth case studies were documented of the junior secondary classes of three teachers trained in different disciplines: biology, chemistry and physics. The three case studies are documented separately: Case study A: Kirkwood, Carr and Gibb, 1987; Case study B: Kirkwood, Carr and Stonyer, 1987; Case study C: Kirkwood, Carr and Shaw, 1987.
•
The documentation of overviews of the data is in Carr and Kirkwood, 1988 and Carr, Kirkwood, Newman and Birdwhistell, 1987a, 1987b. An exploratory primary school study was also done (Poschalk, 1986). Phase 3 consisted of developing strategies and teaching materials to overcome or avoid the documented difficulties in the teaching and learning of energy, using action research. The data is documented in: – – –
the detail of the trial teaching unit (Kirkwood, Carr and Newman, 1988); the use of the pedagogy suggested by the unit (Kirkwood, Carr, Gibb, Shaw and Stonyer, 1988); and the professional development of the three teachers involved in the research (Kirkwood, Carr, Gibb and Stonyer, 1988; Kirkwood and Carr, 1988b).
Findings The themes that emerged from the research on teaching and learning science, started in the first two LISP projects (Learning in Science Projects: Forms 1–4 and Primary), were continued in the third project: the Learning in Science Project (Energy), namely:
An introduction and overview
• • • • • •
9
an investigation into problems and difficulties with teaching and learning ‘energy’; the existing ideas that students and teachers bring to the science lesson; the influence that these existing ideas have on the learning outcomes; teaching that takes into account students’ thinking; pedagogical and curriculum implications of the research findings; and the professional development of teachers of science to enable them to use the research findings.
However, this research is most notable for the determining influence of content on these pedagogical aspects. Hence, the Learning in Science Project (Energy) continued the themes that emerged from the previous two projects. In addition to the ideas students bring to the lesson, the views that teachers bring to their teaching were highlighted (Kirkwood, 1988). A teaching approach that took into account students’ prior ideas and their understandings constructed during the lesson was shown to be more effective than traditional teaching approaches (Kirkwood, 1988). The degree of professional development required by teachers to teach the new approach was considerable (Kirkwood and Carr, 1988b). The recommendations in the final report (Kirkwood and Carr, 1988a) included suggestions for future curriculum development: •
•
•
That the actual concept of energy should not be the focus in primary education (Years 1–8). Rather, students should be provided with many experiences of heat, light, sound, electricity and movement set in familiar contexts. This recommendation is seemingly adopted in the 1993 revision of the primary–secondary science curriculum (Ministry of Education, 1993b: 70–87), in that the term ‘energy’ is not used in the ‘Making Sense of the Physical World’ Achievement Objectives until Level 6 (approximately Years 10–12). The term ‘energy’ is not mentioned at all in the Making Sense of the Material World and Making Sense of the Living World. This is a sharp contrast with the draft 1978 science syllabus (Department of Education, 1978). However, some support material for primary students does address the difference between transfer and the transformation of energy (Hipkins and English, 2000). That in Years 9–11, the energy concept be introduced as a useful and powerful abstraction or model or constructed, invented idea for understanding changes to a wide variety of systems. This recommendation is not evident in the 1993 New Zealand national science curriculum or associated teacher guides. That the concept of energy be taught in the context of defined systems undergoing defined changes. This recommendation is not evident in the 1993 curriculum or associated teacher guides.
10
An introduction and overview
The effectiveness of these recommendations is as yet unresearched. Whilst a document analysis of the current curriculum and guides could be undertaken, research is needed to ascertain the extent and effect of taught and received curricula in science classrooms. The findings of the Learning in Science Project (Energy) are reviewed in Chapters 2 and 4. But what was under-researched were the details of the classroom interaction that resulted in student conceptual development and the details of teacher development process. These were to be researched more fully in the fourth and fifth LISP projects.
Learning in Science Project (Teacher Development) Introduction A thread running through the first three Learning in Science Projects was that of the teacher development to enable teachers to use the research findings to inform their pedagogical practice. This theme, with respect to both pre-service and in-service teacher education, has been ongoing in the research of the University of Waikato staff and students, with respect to, for example: • • • • • • •
early childhood teachers of science (Kirkwood, Bearlin and Hardy, 1989); primary teachers of science (Biddulph, 1987a); pre-service teachers (Biddulph, 1987b, 1990a); using pedagogies based on a constructivist view of learning (Biddulph, 1993); the process of change (Claxton and Carr, 1987, 1991); interviews with teachers (Silvester, 1988a); and the international literature (Silvester, 1988b).
Teacher development to implement teaching approaches based on a constructivist view of learning was a key element of this research. It is paralleled by an interest in teacher development internationally; see for example Hargreaves and Fullan, 1992. Following on from a one-year, exploratory project (Silvester, 1988a, 1988b) the New Zealand Ministry of Education funded a three-year research project, from 1990 to 1992 to investigate ways to promote teacher development in science education. The framework for the project was documented in Bell, Kirkwood and Pearson, 1990a, 1990b and Bell and Pearson, 1991b, 1991c. Teacher development was taken to mean the ongoing process in which teachers are learning, with the main purpose of improving student learning. In particular, the research investigated the teacher development required to help teachers take into account the findings of the three previous Learning in Science Projects, into students’
An introduction and overview
11
learning in science, and the related international research on children’s science, conceptual change and constructivist views of learning. Many teachers, on their own initiative, were wanting to take into account this research in their teaching but had found it difficult to do so (Gilbert, 1993). In addition, the then new science curriculum (Ministry of Education, 1993b) promoted a view of learning and teaching based on the findings of this research into students’ learning. In summary, this research explored the teacher development required for the implementation of new teaching activities (developed as part of the previous LISP projects) by teachers, especially those not involved in the earlier research projects (Leach, Ametler, Hind, Lewis and Scott, 2003). Over the three years (1990–92) the research addressed six questions: 1
To what extent did the teachers on the programmes change their ideas and beliefs about: • • • •
2
3 4 5
6
7
teaching activities and the roles of the teacher; learners and the process of learning, learning-to-learn and teachers as learners; the nature of science and scientific knowledge; science in the school curriculum?
To what extent and in what ways did the teachers change their behaviour in the classroom, with respect to becoming teachers who took into account students’ thinking – that is, constructivist teachers? What factors helped or hindered this development? What resource materials were needed to support this teacher development? What are the strengths and weaknesses of the two kinds of teacher development programmes – the school-based and regionally based programmes? What model of the teacher development can be recommended to policy makers, school managers, teacher educators and teachers as a template for developing programmes? What are appropriate criteria for the evaluation of teacher development courses (Bell, 1993c)?
The Teacher Development Programmes As part of the research, four teacher development programmes were run for a total of 34 teachers of science in the Waikato region in 1989–92. Two of the programmes were school-based and two regionally based. The teachers in the programmes were primary and secondary; women and men; beginning and experienced teachers; and assistant teachers and Heads of Departments. The programmes consisted of after-school sessions over one or two school terms and a small number of classroom visits. The programme sessions comprised sharing
12
An introduction and overview
sessions, in which the teachers talked with each other about the new activities they had tried in their classrooms, and workshop sessions in which the facilitator ran activities to help the teachers reflect on their practice (Bell, 1993b). The programmes run as part of the research are described in detail in Bell and Kirkwood, 1993 and in 1992 the programme, for example, had as its aim to help the teachers • •
• •
develop their ideas of what professional development is and adopt roles for the teacher of learning, researching, supporting and reflecting; develop their classroom practice to take into account students’ thinking, and in particular, adopt new roles for the teacher in the classroom – of listening, managing learning, and responding to students’ thinking; learn about the research findings showing how students learn science; develop a constructivist view of learning and consider its implications for the way we view teaching, the curriculum, knowledge and science.
The 1992 programme, for example, was structured around the teachers considering the roles of a teacher who takes into account students’ thinking. The findings of the Learning in Science Project ( Teacher Development) are reviewed in Chapters 5 and 9. The findings The findings of the Learning in Science Project (Teacher Development) are documented in a book (Bell and Gilbert, 1996) and a summary review (Bell, 1998), including the factors that lead to teacher development, for example support (Bell and Pearson, 1992b), feedback, reflection (Bell and Pearson, 1991a), ‘better learning’ (Bell and Pearson, 1992a), anecdoting (Bell, 1994c); a model of teacher development as personal, professional and social development (Bell and Gilbert, 1994); and a teacher development programme and handbook (Bell, 1993b); an evaluation of a teacher development intervention (Pearson and Bell, 1993) and accounts of teaching that takes into account students’ thinking (Bell, 1993d; Bell, 1994b). Theses on related aspects of teacher education include Cheng, 2000 and Fernandez, 1991. The Learning in Science Project (Teacher Development) concluded at a time when the constructivist view of mind and learning was being strongly critiqued (see Chapter 2 for details). A main criticism of teaching based on a constructivist view of learning was that the teacher was portrayed as a ‘facilitator’ and not a ‘teacher’. The role of the teacher as a provider of information was erroneously seen as teachers not telling and explaining the science to students. But teachers do tell and explain the science to students and in many different ways other than lecturing (Oxenham, 1995). The interaction between teacher and student is not a simple monologue nor characterised by a simple teacher comment and student
An introduction and overview
13
reply. This complex interaction is at the heart of pedagogy and was given more attention in the next and fifth Learning in Science Project.
Learning in Science Project (Assessment) Introduction The teachers in the previous Learning in Science Project on teacher development had indicated that an area of professional development they would welcome is that of interacting with students to discuss and explain science concepts. In taking into account students’ thinking in their teaching, these teachers were responding to and interacting with the students’ thinking that they had elicited in the classroom. A central part of this teaching is dialogue (not a monologue) with students to clarify their existing ideas and to help them construct the scientifically accepted ideas (Scott, 1999). Therefore, giving feedback to students about how their existing conceptions relate to the scientifically accepted ones, and helping them to modify their thinking accordingly, is a part of teaching for conceptual development. It is also a part of formative assessment and formative assessment is seen as a crucial component in teaching for conceptual development (Bell, 1995a; Cowie, 1998). The teachers were therefore undertaking formative assessment whilst teaching for conceptual development. Within this research, formative assessment, as classroom-based assessment for better learning, was defined as ‘the process used by teachers and students to recognise and respond to student learning in order to enhance that learning, during the learning’ (Bell and Cowie, 2001b: 8). This is similar to other definitions in the literature, for example Black, 1993; Gipps, 1994a. Others have used the term ‘formative interaction’ ( Jones, Cowie and Moreland, 2003; Moreland, Jones and Northover, 2001). There has been much written on the role and purpose of formative assessment in improving learning and standards of achievement (Cowie, 1997; Harlen and James, 1996). Black and Wiliam (1998a) reviewed studies that evaluated whether formative assessment did in fact improve learning. Their seminal review of the literature on the impact of formative assessment on learning outcomes boldly stated that The research reported here shows conclusively that formative assessment does improve learning. The gains in achievement appear to be quite considerable, and as noted earlier, amongst the largest ever reported for educational interventions. (Black and Wiliam, 1998a: 61) However, there had been little research on the process of formative assessment itself and the professional development required for its use in classrooms. As Black and Wiliam (1998a) suggest, there is a need to explore views of learning and
14
An introduction and overview
their interrelationship with assessment, and to theorise about formative assessment. In essence, the fifth Learning in Science project took up this challenge. By the mid-1990s in New Zealand, there was an increasing interest in assessment in education, due to the accountability movement in education policy. In particular, formative assessment was increasingly becoming a focus in policy documents on educational assessment and in the professional development of teachers due to its known effects in improving learning outcomes. Research design In this climate of accountability to improve learning outcomes, the fifth Learning in Science Project, LISP (Assessment) was funded by the New Zealand Ministry of Education in 1995–96. It was to investigate assessment in Year 7–10 science classrooms and, in particular, formative assessment (Bell and Cowie, 1997a, b). Formative assessment was being seen as both an effective pedagogy and as a purpose for assessment. The research was based around eight case studies of 10 teachers and their students and the assessment they undertook in the classroom. In addition, the 10 teachers and the researchers met on 11 teacher development days for professional development on classroom-based assessment and to reflect on the data analysis. This enabled a collaborative research method that intentionally combined research and development (Bell and Cowie, 1999; Cowie and Bell, 1995, 1999a). The research did not specifically investigate ongoing or continuous summative assessment, which is also commonly known as teacher assessment, internal assessment, classroom-based assessment and formative assessment. The research aims and findings There were five research aims, all of them adding to our knowledge relating to formative assessment. They were: 1 2 3 4 5
to investigate the nature and purpose of the assessment activities in some science classrooms; to investigate the use of the assessment information by the teacher and the students to improve the students’ learning in science; to investigate the teacher development of teachers with respect to classroom-based assessment, including formative assessment; to investigate the use of assessment information in reporting to parents or others; to develop a model to describe and explain the nature of the formative assessment process in science.
The findings, which are more fully discussed in Chapter 6, included: what was assessed by the teachers and students (Cowie, Boulter and Bell, 1996); the
An introduction and overview
15
purposes for doing formative assessment (Bell and Cowie, 1997a, 1997b); the characteristics of formative assessment (Bell and Cowie, 2001a); a model of formative assessment (Bell and Cowie, 2001a; Cowie and Bell, 1999b); a sociocultural, discursive view of mind and learning (Bell, 2000; Bell and Cowie, 2001b), and teacher development to support the use of formative assessment (Bell and Cowie, 2001c). Aspects of assessment are researched in the theses of Cowie, 2000; Kent, 1996 and Lal, 1991.
Other Waikato research While some of the theses were connected with the Learning in Science Projects as already indicated, others were in the related areas of: • • • • • • • • • • •
teaching and learning of science (Chu, 1997; Oxenham, 1995; Wigglesworth, 1999; Wilkinson, 1993); teaching and learning of biology (Charles, 1988; Raghavan, 1988; Wood, 1996); teaching and learning of chemistry (Rodrigues, 1993b; Standrill, 1983); teaching and learning of physics (Hameed, 1997; Jones, 1982b, 1988, Jones and Osborne, 1985; Mohamed, 1995; Porteous, 1997; Singh, 1993; Taylor, 2000); practical work (Haigh, 1998; Saeed, 1997); Maori and science education (Kent, 1996; McKinley, 1995b; McKinley, 2003); girls and science (Evening, 1998; Gilbert, 1997); curriculum development research (Baimba, 1991; Happs, 1980b; Moli, 1993b; Ume, 1996); tertiary science education (Chang, 2000; Eames, 2003a; Rawaikela Dakuidreketi, 1995); teacher education (Cheng, 2000; Fernandez, 1991); assessment (Lal, 1991).
Most of the research methodologies used by the Learning in Science Projects can be described as within the interpretive paradigm; descriptive, using both qualitative and quantitative data; naturalistic; flexible; triangulated; collaborative; intentionally involving research and development; and ethical. The methodologies are documented in Bell, Osborne and Tasker, 1985; Freyberg and Osborne, 1985; Carr, 1991; Bell, 1993a; Gilbert, 1994; Bell and Cowie, 1997b, 1999; Cowie and Bell, 1999a.
Overview The Waikato research programmes in science education, in the first twenty years, can also be characterised as being research which was:
16
An introduction and overview
• •
underpinned by a commitment to improve the learning of science; addressing two main audiences – national curriculum policy makers (now in the Ministry of Education), and those concerned with the professional development of teachers of science, that is, teachers, teacher educators, advisers (in schools and tertiary institutions). The genre of the research reports and teacher guide material was appropriate to the intended audience; focusing on the learner and learning; evaluating research interventions, for example new teaching and assessment strategies, in terms of learning outcomes; focusing on the teacher, teaching and pedagogy; viewing assessment as an integral part of the teaching and learning of science; linking teaching, learning, assessment and curriculum; undertaken in the main by qualified and experienced teachers; undertaken in primary, secondary and tertiary schools and institutions; undertaken in classrooms; collaborative research with practising teachers and students, acknowledging their subjectivity; in the interpretive research mode; generating both qualitative and quantitative data; eliciting self-report data as well as third party data; using data collection techniques as interviews, surveys, classroom observations.
• • • • • • • • • •
The aim of this book is to give an overview of the research findings of the five Learning in Science Projects and associated theses. The areas of scholarship which have emerged from these findings are: • • • • •
theorising about learning and mind (Chapters 2, 3); pedagogies that take into account students’ thinking (Chapters 4, 5); classroom assessment (Chapter 6); Maori and science education (Chapters 7, 8); curriculum development and teacher development (Chapter 9).
Chapter 2
Constructivist views of learning
17
Constructivist views of learning
Introduction Learning is at the heart of things educational, including science education. The ways in which we view learning can determine the ways in which we approach teaching and assessment in the classroom; the ways in which we think about being a teacher; and how we view and conceptualise students and their needs. All five Learning in Science Projects and associated theses have contributed to the ways in which we view learning. These are outlined in this chapter and Chapter 3, indicating the development of the way learning has been viewed at Waikato over the twenty years: personal constructivist views, social constructivist views, sociocultural views and discursive views of learning. Graham Nuthall (1997), in a review of studies of student thinking in the classroom, identified three broad categories. First, there are those studies (the cognitive constructivist perspective) that incorporate learning and thinking into a broad conception of cognition and students are seen as creating or constructing their own knowledge and skills. A second category (the sociocultural and community-focused perspective) contains those studies that are primarily sociocultural in their orientation. Learning and thinking are seen as social processes or social practices, that is, practices occurring in social contexts – between, rather than within, individuals. The third category (the language-focused perspective) contains studies that have primarily a language or sociolinguistic orientation. ‘Here, the language of the classroom is both the content and the medium of learning and thinking. What students acquire are the lingusitic “genres” of the disciplines’ (Nuthall, 1997: 1). These three categories are evident in the Waikato journey in science education
Alternative conceptions and children’s science The first key finding of the Learning in Science research programmes was that students brought to the science lesson ideas about how and why things behave as they do, that differed from the scientifically accepted ideas. In other words, children’s minds were not blank slates; they already had concepts for such
18
Constructivist views of learning
notions as ‘animal’, ‘force’ and ‘burning’ before encountering formal teaching of science. Much of the research of the in-depth phase of the first Learning in Science project was documenting the nature and extent of these ideas. A total of 13 working papers (Osborne, Freyberg and Tasker, 1980) and subsequent theses and published articles documented the alternative (to the accepted scientific) conceptions held by students on living, animal, plants (Bell, 1981a, b; Stead (now Bell), 1980a), force (Osborne, 1980a); energy (Stead (now Bell), 1980b); particles (Happs, 1980a); friction (Stead and Osborne, 1981a); gravity (Stead and Osborne, 1981b); weather (Moyle, 1980); light (Stead (now Bell) and Osborne, 1979; electric current (Osborne, 1981); physical change (Cosgrove and Osborne, 1980); mixtures (Cosgrove, 1982a) and chemical change (Schollum, 1982a, b). The alternative conceptions held by students coming to the science lesson continued to be researched by masters and doctoral students at Waikato: soils, mountains, glaciers, rocks and minerals (Happs, 1984); photosynthesis (Barker, 1986; Barker and Carr, 1989 a, b, c); chemical stability (Standrill, 1983); genetics (Wood, 1996). Work with international colleagues also included this aspect (for example Gilbert, Watts and Osborne, 1982). The early alternative conceptions work done at Waikato, Leeds, Surrey and Monash Universities soon mushroomed into an international trend in science education research as evidenced by the 2000 or more entries in (Pfundt and Duit, 1994). For example, children’s ideas about ‘animal’ were described and documented (Bell, 1981b; Stead (now Bell), 1980a). In 1980, using an interview-aboutinstances (Osborne and Gilbert, 1979) students aged 10 to 15 were interviewed to establish their concept of animal. Most students, particularly the younger ones, saw animals as mainly the larger land animals, such as those found on a farm, in a zoo, or in the home as pets. The following criteria were used to distinguish between the instances and non-instances of the concept ‘animal’: number of legs, size, habitat and noise production. In contrast, a scientist would use the criteria of heterotrophic feeding, movement, sense organs, nervous systems and cellular structure to distinguish animals from other living things. However, no one of these criteria can be considered as adequate alone, as exceptions exist in all cases. The categorisation of an instance depends on the joint presence of several but not all of the criteria. In summary, of the 39 students interviewed, 35 could not classify all the animal instances as would a scientist, nor know or use the scientifically acceptable criteria of animals. Characteristically, their concepts were a restriction of the scientific one, the range of exemplars being narrower that that of a scientist. Few students categorised a spider, a worm or a butterfly as an animal. These small creatures were often classed as insects because of their size. Only about a half categorised a fish, boy, frog, snail, snake or a whale as an animal. Birds, fish, reptiles and humans, for example, were not seen as subsets of animals but as comparable sets to the set of animal. Of the 150 students (aged 11–18) surveyed, 14 per cent of the Year 7 students, 59 per cent of the tertiary teacher education students and 97 per cent of the first year biology university students correctly categorised all six instances in the survey from the scientific viewpoint.
Constructivist views of learning
19
There are five main findings in this piece of research (Bell, 1993a). First, the students had a concept of ‘animal’. They are not empty vessels waiting to be filled up with scientific knowledge as to what an animal is. They come to science lessons with a concept of ‘animal’, a concept that they had probably developed before formal schooling and which they have been using for many years in communication and in their efforts to make sense of the world around them. Secondly, the concept of animal held by many students is not the concept held by scientists today. That is, the concept of ‘animal’ held by many students differs from the currently scientifically accepted one. The term ‘alternative concept’, rather than ‘misconception’, is used to describe the students’ concept so as to give it value in its own right and not just as an error when compared to the scientific idea. Thirdly, the alternative concept of animal, typically the larger, terrestrial, four-legged, furry creatures, is not just the idiosyncratic concept of one or two students, but is a commonly held belief. Fourthly, the alternative concept of ‘animal’ may be held by students of all ages. Up to ten years of schooling may not have changed their concept of ‘animal’. And lastly, the alternative concept of ‘animal’ held by students is often the everyday meaning of the word used by many adults, including scientists. Even scientists comprehend the sign on the fish and chip shop window saying ‘No animals allowed inside’, using the everyday meaning of ‘animal’, if they are wanting to go inside and buy some food. Further examples of the Waikato research into children’s science is documented in Osborne and Freyberg (1985). Some articles and books from that era that review the accounts of children’s science include Driver (1981, 1983); Driver, Guesné, and Tiberghien (1985), as well as later ones, for example Pfundt and Duit, 1994. 1 Everyday meaning: the only meaning that many students use it is also used by scientists at times 2 Scientific meaning: many students do not use this meaning Animals in a scientific sense person bat fish snake
worm bird weta
Animals in the everyday sense
tree fire car carrot
cow dog cat goat horse
Figure 2.1 The everyday and the scientific meaning of animal
20
Constructivist views of learning
These students’ existing concepts have been labelled as misconceptions, alternative conceptions, alternative frameworks, preconceptions, intuitive ideas or untutored beliefs (Bell, 1993a; Gilbert and Watts, 1983) depending on the emphasis being given by the author. The Waikato group used the terms ‘alternative conceptions’ and ‘children’s science’. ‘Children’s science’ was described as ‘the views of the world and the meanings for words that children tend to acquire before they are formally taught science. Children’s science develops as children attempt to make sense of the world in which they live in terms of their experiences, their current knowledge and their use of language’ (Osborne, Bell and Gilbert, 1983: 1). In addition, the term ‘children’s science’ emphasised ‘both the similarity of the process of construction of meaning by children and scientists and the differences in the outcomes of that construction’ (Gunstone, 1988: 88). The research on alternative conceptions and children’s science was considered important by the Waikato team for several reasons. •
•
•
The research indicated the importance of children’s science in determining the learning outcome. For example, a study reported by Bell and Barker (1982) suggested that the difficulties students experience with abstract ecological concepts (such as ‘consumer’) may be in part due to students not having the scientifically accepted concepts of more basic ideas such as ‘animal’. Two classes of 13-year-old students were surveyed about their concepts of ‘animal’ and ‘consumer’ before and after a teaching episode. The control class was taught by a teacher who had no knowledge about children’s ideas of animal and the teaching consisted of the lessons suggested for the ecological study in the school scheme. It included work on the ideas of ‘habitat’, ‘producer’, ‘consumer’, and ‘decomposer’. The experimental class undertook activities which helped them learn the scientific concept of ‘animal’ and only at the end of the lesson was the concept of ‘animal’ related to the concept of ‘consumer’. The results indicated that the control group did not have scientifically acceptable concepts of ‘animal’ or ‘consumer’ before or after the teaching episode. The results of the experimental class before the teaching episode were similar to those of the control class. However, the postteaching results indicated that the students had learnt the scientific concept of ‘animal’ and ‘consumer’. In other words, the teaching that took into account students’ existing ideas of ‘animal’ was more effective in helping students learn the scientific concept of ‘consumer’. Further examples are given in Osborne and Freyberg (1985). The research described the unintended and unanticipated changes in students’ ideas. When students’ ideas are changed by teaching, they may be changed in ways unanticipated or unintended by the teacher (Gilbert, Osborne and Fensham,1982; Osborne, Bell and Gilbert, 1983). The work of Ross Tasker (1981) gives us some insight into why teaching may not have been as effective as we thought. He found, through naturalistic observation, that in many science classrooms:
Constructivist views of learning
(i)
(ii) (iii)
(iv)
(v)
•
21
pupils tended to consider each lesson as an isolated event while the teacher assumed that the pupils appreciated the connecting link between the lesson and the previous learning experiences. pupils sometimes invented a purpose for the lesson which was subtly but significantly different from the purpose intended by the teacher. pupils often showed little interest in, or concern about, those features of an investigation which the teacher, or textbook writer, considered to be critical scientific design features. pupils’ knowledge structures, against which learning experiences were considered, were frequently not the structures the teacher assumed pupils had. pupils’ understandings, developed from the outcomes of experimental work, were frequently not those that the teacher assumed were developed (Tasker, cited in Osborne and Wittrock, 1983: 490–1).
One other finding from this work (Tasker, 1981) is that in situations where pupils were involved in teacher- or textbook-guided investigations, pupils spent much of their time making executive decisions (What do we do now? What instruction are we up to?) and very little time really thinking about concepts in science. In other words, the students were not engaging with the ideas of science; that is, ‘minds-on’ science education was not occurring. An important discussion and critique of the notion of ‘children’s science’ was in comparing it to scientists’ science (Bell, 1993a), and in particular the learner-as-scientist (Driver, 1993); ‘man (sic)-the-scientist (Kelly, 1955); the differences in the views generated by children and those of scientists (Gilbert, Osborne and Fensham,1982; Osborne, Bell and Gilbert,1983); the parallels between the ideas of students and historical scientific ideas (Posner, Strike, Hewson and Gertzog, 1982; Nussbaum, 1989); and the purpose of meaning making (Hills, 1989).
Learning as conceptual development The research findings on alternative conceptions and children’s science led the Waikato group to develop their theorising on learning, in terms of children constructing understandings and developing their existing ideas, that is, learning as conceptual change or development. Some accounts of a constructivist view of learning written for teacher audiences may be found in Driver and Bell, 1986 and Gunstone, 1990. More detailed accounts of constructivism from the 1980s can be found in (Driver and Erickson, 1983; Pope and Gilbert, 1983; Wheatley, 1991). Students’ prior knowledge had been considered important for many years in the teaching and learning process. Bruner (1977: ix) stated that ‘one starts somewhere – where the learner is’ and Ausubel (1968: vi) ‘the most important
22
Constructivist views of learning
single factor influencing learning is what the learner already knows. Ascertain this, and teach him accordingly’. But the notion of existing knowledge in this previous theorising was whether the students had the ‘correct’ knowledge or not, and little attention was given to the nature and significance of the students’ ‘incorrect’ knowledge. Students come to science lessons with their own ideas about phenomena, meanings for words and explanations of why things behave the way they do. Learning, therefore, is not about filling students’ empty heads or students acquiring new ideas per se, but about students developing or changing their existing ideas. Learning is seen as conceptual change, the construction and acceptance of new ideas or the restructuring of existing ideas. This view of learning, called the constructivist view of learning, recognises that students construct rather than absorb new ideas and that learners actively generate meaning from experience – both physical and social experiences. A model of conceptual change was developed by Posner et al., 1982, expanded by Hewson (1981), and reviewed by Hewson and Thorley (1989). It is based on an analogy between conceptual change in the scientific disciplines and conceptual change in people learning science. Two major components of this model of conceptual change are the ‘conditions that need to be satisfied in order for a person to experience conceptual change and the person’s conceptual ecology that provides the context in which the conceptual change occurs and has meaning’ (Hewson and Thorley, 1989). The four conditions, which need to be seen by the learner as being met before conceptual change will occur, are whether the new conception is intelligible, plausible, and fruitful for the learner and whether the existing conception is a source of dissatisfaction for the learner: 1
2
3
Is the conception intelligible to the learner? Does the learner know what it means? Do the pieces of the conception fit together for the learner? Is the learner able to find a way of representing the conception? Can the learner explore the possibilities inherent in it? Is the conception plausible to the learner? If the conception is intelligible to the learner, does the learner also believe that it is true? Is it consistent with and able to be reconciled with other conceptions accepted by the learner? Does the new conception make sense to the learner? Is the new conception fruitful for the learner? If the new conception is intelligible to the learner, does the learner also find that it achieves something of value for him or her? Does it solve otherwise insoluble problems for him or her? Does it suggest new possibilities, directions, ideas? The extent to which the conception meets these three conditions is termed the status of a person’s conception. Expressed in these terms, the conceptual change model is about changing, i.e. raising or lowering, the status of conceptions. As more conditions are met, the conception’s status is raised.
Constructivist views of learning
4
23
Intelligibility is a necessary first step in raising status. Without intelligibility, a conception has no status to a person and cannot become either plausible or fruitful. If conditions that were once met are no longer seen to be, the conception’s status is lowered. It is, however, unlikely that a conception will become unintelligible, i.e. once we know what an idea is, even if we no longer believe it, we are likely to remember it. The fourth condition in the conceptual change model is directly related to the changes in status. Is the conception a source of dissatisfaction to the learner? Does the conception seem to be counter-intuitive to the learner? That is, does it seem that it is not plausible to the learner? Does it create difficulties or block possibilities for the learner? That is, does it seem that it is not fruitful for the learner? (Hewson and Thorley, 1989)
When a learner considers a new conception, two possibilities exist, according to Hewson and Thorley (1989). It can be incorporated with existing conceptions in a process called assimilation (Posner et al., 1982) or conceptual capture (Hewson, 1981). Or the new conception can be accepted, lowering the status of the former (and conflicting) conception and raising the status of the new conception in a process called accommodation (Posner et al., 1982) or conceptual exchange (Hewson, 1981). Both the processes of conceptual capture and conceptual exchange are examples of conceptual change. In summary, the process of learning as conceptual change is seen as ‘an adaptive process, one in which the learners’ conceptual schemes are progressively reconstructed so that they are in keeping with a continually wider range of experiences and ideas. It is also seen as an active process of ‘sense making over which the learner has some control’ (Driver, 1989). The learner is actively involved in knowledge construction (Claxton, 1984, 1991). The notion of ‘active’ with respect to knowledge construction was further described by Bentley and Watts (1989) and Bell and Pearson (1992a), as when the students, for example: •
•
• •
have ownership of their own learning. They are not learning because of a syllabus or prescription but to satisfy their own desire to ‘make sense of their world’ or to resolve some confusion or conflict. This is often expressed as students taking responsibility for their own learning; are making links for themselves between what they already know and the new ideas. The students see this as a part of learning and they have the expectation that they will have to do this in order to learn; are constructing new ideas and testing them out, to judge the worth of ideas and opinions; are motivated to learn from within. There is little need for teacher-initiated stimulus material or experiences after the unit of work has begun. The students are enjoying their learning, despite the thinking being ‘hard work’ at times;
24
Constructivist views of learning
•
want to know what other students (and the teacher) are thinking and investigating; feel good about themselves. They are confident to voice their own ideas and justify them in a small or whole-class group; are coping with a range of feelings. Learning involves taking risks and moving away from the controlled, comfortable, predictable, and known. Learning feelings are those of being anxious, insecure, threatened, surprised, shocked, disappointed, confused, frustrated, irritable, distressed, fearful, hopeful, excited, intellectually satisfied. These feelings are an integral part of learning and cannot be removed entirely. Students will be seeking support from each other and the teacher in terms of coping with these feelings; are using new ideas within familiar and new contexts until they can use them with confidence.
• •
•
Parallel to this work on conceptual change and conceptual development was theorising of learning science developed under the broad heading of ‘constructivism’, including the Waikato research on the Generative Learning Model (Osborne and Wittrock, 1983, 1985). The following review of constructivism (adapted from Bell and Gilbert, 1996) places the Waikato research in the international debates in the 1980s and 1990s. Taken in its most general form, constructivism asserts that all learning takes place when an individual constructs a mental representation of an object, event or idea. Mental representations are used as a basis for mental and physical action, and both enable and constrain an individual’s process of meaning making (Resnick, 1991). These mental representations may be referred to as knowledge or beliefs, with distinctions made between them on the basis of the criteria of justification. However, the terms ‘knowledge’ and ‘belief’ are here used interchangeably, with the terms ‘public’ knowledge and ‘individual’ knowledge being used as well. Whilst such a definition of constructivism does enable a wide diversity of interpretations of learning to be gathered under one label, its looseness can inhibit theorising on the learning process. Three major sub-groupings within constructivism can be distinguished in the early work of the Waikato group: Piaget’s (1970) approach; Kelly’s (1969) personal construct psychology approach; and personal constructivism as exemplified by Osborne and Wittrock (1985). A brief exposition of each is given, as well as a review of the criticisms.
Piagetian constructivism Piaget’s work was to challenge the then behaviouristic and hierarchical views of learning science (White, 1974), which had resulted in teaching and curricula being structured around the hierarchical and logical structure of the discipline (Bruner, 1977) as a way to sequence the content to be taught and learnt. The teaching sequence (and by implication the best learning sequence) was that determined by the structure of the discipline. The simple and less complex
Constructivist views of learning
25
ideas of the discipline could determine a hierarchical structure of basic to more complex ideas to be taught (and learnt). But this view began to be challenged, for example, by Raven (1967–68), who challenged the logical and hierarchical analysis of the momentum construct of: ‘conservation of matter → speed → proportional use of mass and speed, with momentum held constant → momentum’, as the sequence in which students could best learn the concept of momentum. The research suggested that the conceptual sequence of ‘momentum → conservation of matter → proportional use of mass and speed, with momentum held constant → speed’ better described the learning sequence for children. The early work on ‘force’ by the first Learning in Science Project (Osborne, 1980b) reinforced this challenge. The research on children’s science challenged science educators to structure curricula and teaching, starting with the ideas children brought to the science lesson, and not necessarily the subordinate and less complex ideas of the discipline. This has been a major challenge and one that twenty years later is still challenging teachers and curriculum developers as they struggle to promote learning in science in an educational culture dominated by accountability demands for indications of progression in learning. Piaget’s approach proposes that a person’s mental representations are produced during progressively more complex interactions by that individual with the world of physical objects. Incoming information is initially assimilated by existing mental structures. If this assimilation proves inadequate – that is, the incoming material cannot be understood in terms of the existing mental structure, accommodation takes place – that is, a modified structure evolves. The interplay between assimilation and accommodation, known as equilibration, results in mental structures which are progressively more decentred – that is, are less and less concerned with the immediate, the concrete and the personal. In terms of pedagogy, Piaget’s work encouraged science educators to take ‘an active view’ of learning and attempting to match the types of experiences given to children to the general pattern of cognitive growth outlined by Piaget – working with concrete objects for younger children and progressing to ideas requiring more formalisation later (Driver, 1982). In terms of content, Paiget’s work was used to design curricula that took into account the stage of cognitive development of the majority of students with respect to the sequencing of content and the nature of the content itself (Shayer and Adey, 1981). For example, the Australian Science Education Project (ASEP) (Power, 1972), was written taking into account the notion that many children in the lower secondary school would not have attained Piaget’s stage of formal operations. At the time, existing curricula were largely based on the logical structure of the discipline, in which, for example, it was suggested that the concept of ‘energy’ should be taught to young students as it was a basic concept in science. From the Piagetian perspective, the concept was seen as too abstract for the majority of students to learn until they had reached the stage of formal operations, that is, until about 15–16 years. Calls were made for the revision of curricula. However, the danger with this belief is that the curriculum becomes
26
Constructivist views of learning
impoverished – ‘the students are not ready for this work yet’, and teaching may be viewed as making no difference to children’s learning, since maturation is seen as the major influence on whether learning will occur or not. In addition, Piaget’s work had influenced the content of some curricula by reflecting the Piagetian ‘schemes’, for example, classification, ordering, the use of controls and variables. These were seen by science educators as important in terms of scientific enquiry and the development of these enquiry skills became the curricula objectives themselves. Furthermore, the acquisition of these skills was viewed as being context-independent. Once acquired, they were believed to be transferable to any context. Whilst it is primarily concerned with the development of scientific rationality, and therefore perhaps of special significance to the learning of science teachers, it is a stage theory, and therefore subject to all the weaknesses outlined in Driver (1978) and Burden (1990). In addition, the notions of stages of development and structures were being challenged. Carey (1986: 1129) summarised that many developmental psychologists now believe that the young child does not think differently from the adult, is not concrete, illogical and so forth. Phenomena that were interpreted in terms of Piaget’s stage theory are better interpreted in terms of specific alternative conceptual frameworks – novice– expert shifts and theory changes in particular domains. There are a number of other problems with Piaget’s approach to constructivism (O’Loughlin, 1992). First, Piaget’s model assumes that an individual comes to understand the world as it is; that is, comes to know reality in order to adapt to it. This realist approach is conservative in outlook, in that individuals can only converge on one ultimate mental structure, that of so-called formal operations which operate on a world which cannot be changed. Secondly, construction for Piaget refers to ‘the process of constructing abstract, decentred, content-free, representations that are universal enough to be modelled by mathematical formalisms’ (O’Loughlin, 1992: 795). Development is seen as the development of content-free logical structures and operations. This view of construction ignores the socially and historically situated nature of knowing. It gives ‘primacy to abstract mental structures and rational thought processes at the expense of the historically and socially constituted subjectivity that learners bring to the reasoning process’ (O’Loughlin, 1992: 800). Thirdly, the model infers that communication is only possible between individuals within the limits set by the capabilities of the person at the lower stage of development. Lastly, the process of knowledge construction is seen as individual and personal, with no attention being given to the social. As O’Loughlin (1992) puts it after reviewing a range of critics of Piaget’s theory: . . . knowledge is socially constructed . . . we cannot talk of knowing without considering the historically and socially constituted self that engages in the
Constructivist views of learning
27
process of knowing . . . knowing is a dialectical process that takes place in specific economic, social, cultural, and historical contexts. Knowing is . . . a process of examining current reality critically and constructing critical visions of present reality and of other possible realities so that one can become empowered to envisage and enact social transformation. (O’Loughlin, 1992: 799) Although Piaget has articulated a social dimension of learning (Chapman, 1986; Mays, 1982), this dimension is not as visible in accounts and critiques of his work as the individual dimension. Critiques of the work of Piaget and education may be found in Brown and Desforges (1977), Donaldson (1978), Driver (1978, 1982), Driver and Easley (1978), Novak (1978) and White (1988). The dissatisfaction with stage aspects and epistemological aspects of his work by the educational community led to other views of conceptual development being explored.
Personal construct theory The second form of constructivism found in the Waikato work is that of Kelly (1955). Kelly’s personal construct psychology (see also Claxton, 1984; Pope and Gilbert, 1983; Pope and Keen, 1981) leaves the issue of realism to one side: ‘the open question for man (sic) is not whether reality exists or not but what he (sic) can make of it’ (Kelly, 1969: 25). Through his phrase ‘the person can be seen as the scientist’ (Pope and Keen, 1981: 26). Kelly proposed that each person constructs a representational model of the world, composed of a series of interrelated personal constructs, or tentative hypotheses about the world, with which past experience is described and explained and future events are forecast. Communication is possible to the extent that one person can construe, or understand, another person’s construct system; a similarity of construct systems is not strictly necessary. In this sense, Kelly’s constructivism is also implicitly social in that, as a clinical psychologist, he was concerned with the relationships between people and especially how each individual construed them. Kelly’s great contribution to constructivism is his assertion that there are no predetermined limits on constructs in terms of the nature and range of their application. The limit to their creation is only set by the imagination of the individual concerned and by the constructs being continually tested for their predictive and explanatory adequacy in physical and social contexts: those that prove successful will be retained, used again, and used in a wider range of contexts, whilst those that do not will be modified or abandoned. The apparent weakness of his theory is the lack of emphasis on the impact of others on the production, testing and modification of a person’s constructs, that is, the sociocultural aspects of learning are little considered. Also, a very high level of autonomy of agency is assumed, that is, the person is able to make changes to herself or himself readily. Kelly’s constructivism and theorising was used within the Waikato research by Happs (1984) and Stead (1984).
28
Constructivist views of learning
These first two sub-groups of constructivism (Piaget and Kelly’s theorising), which were seen as focusing on the individual and on the personal construction of meaning, informed the research into children’s learning in science internationally in the 1980s, including research in Australasia (Northfield and Symington, 1991) and Europe (Driver, 1989; Driver, Guesné and Tiberghien, 1985), and the Learning in Science Projects in New Zealand (for example, Bell, 1993a; Osborne and Freyberg, 1985). In particular, the role of prior knowledge in learning was considered, as was the domain-specific nature of learning and the conceptual development view of learning (Gilbert, Osborne and Fensham, 1982; Osborne, Bell and Gilbert, 1983).
Personal constructivist view of learning In New Zealand, a personal constructivist view of learning (arising from cognitive psychology) was developed by the researchers at the University of Waikato to theorise the research findings on alternative conceptions and ‘children’s science’ and was best articulated in the Generative Learning Approach (Osborne and Wittrock, 1983, 1985; Osborne and Freyberg, 1985). Generative learning was described as follows: The fundamental premise of generative learning is that people tend to generate perceptions and meanings that are consistent with their prior learning. These perceptions and meanings are something additional to both the stimuli and the learner’s existing knowledge. To construct meaning requires effort on the part of the learner and links must be generated between stimuli and stored information . . . (Osborne and Wittrock, 1985: 65) Its foundations in information-processing, cognitive science and psychological views of cognition were evident: The generative learning model is centrally concerned with the influence of existing ideas on what sensory input is selected and given attention, the links that are generated between stimuli and aspects of memory store, the construction of meaning from sensory input and information retrieved from long term memory, and finally the evaluation and possible subsumption of constructed meanings. (Osborne and Wittrock, 1985: 65) The key postulates on which this constructivist view of learning is based were given as follows: (i)
The learner’s existing ideas influence what use is made of the senses and in this way the brain can be said to actively select sensory input.
Constructivist views of learning
29
(ii)
The learner’s existing ideas will influence what sensory input is attended to and what is ignored. (iii) The input selected or attended to by the learner, of itself, has no inherent meaning. (iv) The learner generates links between the input selected and attended to and parts of memory store. (v) The learner uses the links generated and the sensory input to actively construct meaning. (vi) The learner may test the constructed meaning against other aspects of memory store and against meanings constructed as a result of other sensory input. (vii) The learner may subsume constructions into memory store. (viii) The need to generate links and to actively construct, test out and subsume meanings requires individuals to accept major responsibility for their own learning. (Osborne and Wittrock, 1985: 65–7)
The personal but not the social construction of meaning was considered and individuals were seen as being able to change their own thoughts and actions. There was no explicit acknowledgement of the sociocultural perspectives of learning as there was no acknowledgement of the affective and intentional aspects of thinking and learning (Bell, 1984a; Gilbert, 1997; Pintrich, Marx and Boyle, 1993). This personal constructivist view of learning (often with acknowledgement to the social) and conceptual change view of learning was used to underpin other research and development at the University of Waikato, for example Porteous, 1997; Raghavan, 1988; Saeed, 1997; Taylor, 2000. In summary, the main points of the 1980s personal constructivist view of learning can be summarised as follows: 1 People construct their own interpretations of communications and experiences. 2 Personal interpretation is determined largely by existing beliefs, which are prior constructions. 3 Interpretation is often influenced, although not necessarily determined, by the interpretations expressed by others – parents, teachers, peers, texts, other media. 4 Students at all levels enter the classroom already holding beliefs relevant to the topic. 5 The extent of beliefs and the intensity with which they are held varies from topic to topic. 6 There will be a range of beliefs among the students. 7 Students’ beliefs about scientific principles and natural phenomena often differ from the scientists’ established views.
30
Constructivist views of learning
8 Where students’ views differ from those of the scientists, they are less economical in interpreting or predicting outcomes of an extended range of events. 9 A person can hold beliefs that contradict each other, applying one belief in one context, another belief in another context. 10 People often interpret events in a manner that supports a belief, and so avoid confronting discomforting instances. That is, they see what they believe. 11 People might alter their memory of an event that contradicts a belief so that their recollection is consistent with what they believe. 12 Beliefs resist change, but students can exchange an alternative conception for the scientists’ conception. 13 Contradictory beliefs can be resolved. 14 Change of belief, or resolution of contradictions, is usually slow and requires repeated experiences favouring the final accepted interpretation. 15 Teaching that encourages resolution of alternative conceptions with the scientists’ view will include elucidation of the students’ beliefs, discussion of the beliefs and their implications, and the design and execution of events that test the accuracy of the beliefs. (Adapted from White, 1991) Further summaries of the Australian and New Zealand research into children’s science, constructivism and learning in science can be found in Bell (1993a) and Northfield and Symington (1991).
The importance of context The context in which the learning occurred was seen as important in the views of learning developed by the Waikato group in the 1980s and 1990s. The generative learning model (Osborne and Wittrock, 1985) described the links made between the incoming stimuli and the existing knowledge of the student, with the context of the learning mediating what new knowledge or learning each student constructed. This role of context in learning was addressed in the research of Gribble, 1993; Jones, 1988; Jones and Kirk, 1989, 1990a, b; Mohamed, 1995; Porteous, 1997; Rodrigues, 1993b; Rodrigues and Bell, 1995, as well as in discussions of schema theory (Bell, 1984a; Cosgrove, 1989; Singh, 1993; Singh and Carr, 1992). Context was also being addressed in the international literature, in terms of conceptual ecology (Posner et al., 1982), the transfer and non-transfer of learning from one context to another (Bell and Brook, 1984; Millar and Driver, 1987; Perkins and Salomon, 1989) and general cognitive skills and domainspecific skills (Adey, 1997). However, the debates on context at Waikato gave rise to the further development of views of learning to address social and cultural contexts of learning. Learning was seen as not being context-free, as it is embedded in a social and cultural context (Bruner and Haste, 1987; Vygotsky, 1978). As Bruner (1990) said:
Constructivist views of learning
31
it is culture . . . that shapes human life and the human mind, that gives meaning to action. It does this by imposing the patterns inherent in the culture’s symbolic systems – its language and discourse modes, the forms of logical and narrative explication, and the patterns of mutually dependent communal life. (Bruner, 1990: 34) The context of learning makes knowledge possible and actions meaningful (Hennessy, 1993).
Critiques of constructivism Over these first twenty years of the research in science education at Waikato, constructivism was increasingly adopted internationally by researchers, curriculum developers and teachers in science education as a view of learning and knowing by students and teachers. This consensus was reached, to some extent, by avoiding a debate about its central ideas (Suchting, 1992). Many of those who have contributed to the development of a constructivist approach to science education (including the Waikato group) agreed with Solomon (1994) that there was a need ‘. . . to try to avert a long period of stalemate while an over-used theory slides into decline’ (p. 17). The following review of and response to the critiques of constructivism is based on that found in Bell and Gilbert (1996: 51–8). Constructivism has provided a powerful and fruitful research programme in learning in science education (Duit, 1994). However, the plethora of interpretations of constructivism, and the foundation of much of the work in the field on an amalgam of them which has not been defined expressly, have led to the emergence of a number of criticisms of constructivism per se. Criticisms have come from both those within and those outside the broad field of constructivism (as discussed throughout this chapter) and chiefly from a philosophical perspective (Gilbert, 1997; Matthews, 1998). The main criticisms concern the loosely defined terms; the search for the ‘grand theory’; ontological and epistemological concens; and a constructivist view of knowledge. These main criticisms are now discussed. Loosely defined terms One of the main criticisms of constructivism has been the use of loosely defined terms, for example, ‘active’, ‘engaged’ and ‘construction’ (Jenkins, 2000; Nola, 1995). However, as Duit (1994) points out, this so called looseness or vagueness has also been a strength, for it has allowed a creative development of thinking within the broad frame of constructivism, which would not have been possible in a more closed theory, with precise definitions. More well-defined use of terms is now sought by those working within and outside the field of constructivism to
32
Constructivist views of learning
prevent this powerful and useful view of learning being thrown out ‘with the bath water’. The search for the ‘grand theory’ Constructivism has been critiqued as a partial account of learning and knowing; it is a cognitive approach to learning, based on a rational style of thought only (McComish, 1994). It has little to say about the affective aspects of learning, nonrational thinking and skill learning, or about culture and power in the classroom (O’Loughlin, 1992). This is one of the biggest challenges to constructivism from those working within its framework – that it does not address the non-rational, affective and intuitive aspects of learning. However, the assumption that there is something called a ‘grand theory of learning’ yet to be invented needs to be challenged (Gilbert, 1997) as did Solomon (1994: 17), when she stated that ‘to equate the absence of such total coverage with theoretical error illustrates once again the overblown expectations that have accrued to constructivism’. It could well be that there will be several theories of learning, each giving a partial view. Post-structuralist and discursive views of science education and learning are developing possibilities (Gilbert, 1997) as discussed in Chapter 3. Ontological and epistemological concerns Jane Gilbert, in her doctoral thesis stated: For philosophers, there have traditionally been two central questions through which the problem of defining knowledge has been conceptualized. These are the problem of distinguishing ‘knowledge’ from ‘belief’; and the (related) problem of developing an adequate basis on which ‘true’ knowledge can be defined and justified . . . science is conceptualized by scientists as true knowledge which is justified by checking it against its origins in the ‘real’ world. (Gilbert, 1997: 177–8) Hence critiques of constructivism by some philosophers, for example Matthews (1994), focus on the difficulties with validating constructed knowledge in terms of ‘truth’, in science and in science education for example. Are the scientists’ or students’ constructions knowledge or belief? How do scientists and students know if the constructed knowledge is ‘true’? Therefore, the two main philosophical criticisms of constructivism have been that of its view of the nature of knowledge, and in particular, of scientific knowledge, and how we come to know this knowledge. Jane Gilbert has succinctly reviewed these philosophical criticisms of constructivism, identifying the key issue of the early views of constructivism in science education as being the lack of development of ‘an account of the nature of scientific knowledge which is
Constructivist views of learning
33
compatible with the theories of learning being advocated (Gilbert, 1997: 218). She argued that science educators need to distinguish between the development (and justification) of new knowledge and the learning of existing knowledge. She further argued that this could be done by two strategies: One involves the development of a clear distinction between accounts of how a given body of knowledge came about and how it can be justified, and accounts of how people might learn that body of knowledge (see, for example, Osborne, 1996; McMillan, 1995). The other strategy that has been suggested involves the development of conceptions of the nature of learning and of science that are theoretically compatible (see, for example, Carr et al., 1994; Driver, Asoko et al., 1994). Both these strategies however avoid dealing explicitly with the question of the status of the students’ subjectivity with respect to scientific knowledge itself, and both maintain existing conceptions, not only of scientific knowledge, but, more generally, of rationality, individuality and subjectivity. (Gilbert, 1997: 352) Because of this perceived inability (as claimed in the critiques) of constructivism to account for the validation of knowledge as ‘true’, and constructivism’s view of viable rather than validated knowledge, the constructivist perspective is sometimes critiqued as involving a relativist view of knowledge, that is, there is no absolute truth, only socially constructed relative truths. The issue of relativism is currently being debated and argued in science and science education, and it is largely unacceptable to many in science education, given their science backgrounds with the commitment to experimentation. Other philosophical positions are possible, for example Yeatman (1994, as cited in Gilbert, 1995), in describing an epistemological basis of contemporary postmodern feminist theory, noted that ‘the perspectivalist theory of knowledge is relational not relativist’ (p. 15). And, more importantly, if, as feminist philosophers have argued, the pursuit of absolute truth does not necessarily go hand in hand with an end to domination, then it is logical that, for these feminist philosophers, there will be a move away from a concern for absolute truth, and a move towards a closer examination of the assumptions which underlie the very possibility of such a concept. (Gilbert, 1995: 14) Gilbert herself (1997: 351) argued for an explicit refusal to ‘resolve’ what ‘Matthews (1994) characterises as “wider epistemological concerns” in science education’. And a socially constructed view of scientific knowledge and students’ scientific knowledge need not imply relativism – a realist position can be adopted (Carr et al., 1994; Driver, Asoko et al., 1994; Leach and Scott, 2003; Osborne, 1996). A relativist position can of course be adopted as in Gilbert (1997), as explained
34
Constructivist views of learning
in the previous section. But all the realist, relational and relativist positions are alternatives to empirical and positivist views. Within western society, scientists and philosophers, and religious fundamentalists (who have argued their definition of the Truth as ‘word of God’), have all laid claim that their discipline is the key one concerned with the arbitration of knowledge and truth. Their modernist project appeared to be to articulate a single, universal theory of knowledge and truth. But these three groups have been challenged by other groups also claiming to be arbitrators of knowledge and truth, making visible the power/knowledge nexus. In the postmodern project, these other groups are claiming a reality based on their lived experiences, for example, women, Afro-Americans and indigenous peoples, including Maori, within other disciplines, for example, the social sciences, including education, sociology and psychology. It was within this postmodern context that the critiques of constructivism were also placed. Another key aspect of constructivism is the active construction of knowledge (Olssen, 1995a), which can be described as: ‘rather than searching for a correspondence or match between mind and the world, the constructivist searches for functional adaptation or fit between new knowledge and prior experience, placing a much greater emphasis on the ability of the human mind to construct and impose categories on the world’ (Olssen, 1995a: 50). Hence, a constructivist approach to science education emphasises ‘making sense’: both making sense of the everyday world and making sense of science, and viability of knowledge; both making sense of public, socially shared knowledge and making sense of new personally constructed knowledge (Millar and Driver, 1987). There is also a discussion as to how the initial experience or prerequisite knowledge is acquired if it has not been experienced (Osborne, 1996) and the role of genetic information, reflexes and the automatic nervous system. A personal or individual approach to constructivism runs a risk that it might be inferred as promoting a view that science and learning proceed by an empirical– inductivist route (Matthews, 1992). The critique is that constructivism is empiricist, particularly if it is individual-centred and experience-based (Matthews, 1992). A constructivist view of learning does not advocate this (Leach and Scott, 2003). ‘Making sense’ is a phrase used to indicate the reconstruction of already known public/social knowledge as new personal knowledge by the students as part of their learning. The inputs into the construction of knowledge may be those sensed by touch and sight in terms of observations of the physical world but they may also be inputs from the social world sensed by sight and hearing during communication. Moreover, the individual-centred, experience-based, empiricist label is difficult to sustain if social constructivist, sociocultural and discursive positions are adopted (Driver, Asoko et al., 1994; Olssen, 1995b). These views are discussed in Chapter 3. Both these ontological and epistemological criticisms related to empiricism and relativism are sometimes linked to curriculum debates with the misleading
Constructivist views of learning
35
statement that ‘anything goes’ and students will not learn the content in a state or national curriculum if they are ‘making sense for themselves’, implying that the students are not learning the currently accepted ideas of scientists. However, teachers who view learning and teaching from a constructivist position are morally and (often) legally committed to helping students to construct meaningful understandings for themselves of the scientific concepts in the curriculum. A view of learning from a constructivist perspective is not the same as discovery learning – the teacher intervenes when necessary to engage with the thinking of the students (Cowie and Bell, 1999b; Oxenham, 1995) to promote the learning of currently accepted scientific knowledge.
A constructivist view of knowledge A related critique of the constructivist view of learning is its view of knowledge, for within science education, constructivist views of learning have challenged the way scientific knowledge is seen. As Carr et al. (1994) noted: Many teachers hold the view that: • • • •
science knowledge is unproblematic; science provides the right answers; truths in science are discovered by observing and experimenting; the choices between correct and incorrect interpretations of the world are based on commonsense responses to objective data.
This traditional image of science has been explored in a large number of commentaries; see for example, Chalmers, 1976; Nadeau and Desautels, 1984. (Carr et al., 1994: 147) Roth and Roychoudhury (1994) also stated: At present, most science teaching is based on an objectivist view of knowing and learning . . . (which) subsumes all of those theories of knowledge that hold that the truth value of propositions can be tested empirically in the natural world . . . traditional science teaching has focused on the direct transmission of these truths. (Roth and Roychoudhury, 1994: 6) It was argued that this view of scientific knowledge leads to so-called traditional approaches to teaching science, which stress the transmission and acquisition of a body of currently accepted scientific knowledge, in an orderly way based on the nature of the discipline. In this view, knowledge is seen as real, true and factual.
36
Constructivist views of learning
However, in constructivist views of learning, knowledge is seen as being constructed by scientists (public, socially constructed knowledge) and reconstructed by each learner of science (personal individually constructed knowledge). Reality is recognised as being independent of cognising beings, but something that cannot be ‘known’ directly. It questions ‘whether or not scientific laws are coextensive with the “real” world, and/or whether or not the real, physical world is – or can be – adequately “known” . . .’ (Gilbert, 1995: 14). For example, the concepts of ‘energy’, ‘floating’ and ‘sinking’, and ‘photosynthesis’ do not exist in nature but are ideas invented (constructed) by scientists to help them better make sense of the world (Carr et al., 1994). Meaning is not seen to exist in the natural world. Scientific knowledge is not seen as a one-to-one reflection of or correspondence with the ‘real world’ but it is about the ideas, concepts and theories constructed by cognising beings and used to explain that world. Knowledge is seen as ‘viable’ in the experiential world rather than ‘true’ and ‘absolute’. An early quote from Driver and Bell (1986) further illustrates a constructivist view of scientific knowledge: Consider a simple example: a plastic comb is rubbed with a cloth and it then picks up some small pieces of paper. In this case the events in the ‘real world’, the world of sense impressions (i.e. the world of objects and events we can point at, touch and see) include the rubbing of the comb and the pieces of paper being picked up. Yet consider for a moment the kind of explanation we might present for this in a science lesson; an explanation which might involve ideas about charge transfer from the comb to the cloth, net charge on the comb producing an electric field, the field causing distortion of electron clouds in the paper and hence a net force on the paper. The entities we are talking about here, electrons, electric fields, etc., are not part of the world of sense impressions, they are not even abstracted from this world. They are imaginative constructions, themselves related in very precise ways, which are brought to bear on this world . . . The other point concerns the nature of the relationship between sense impressions and theory. Concepts and theories do not follow from observation in a simple ‘inductivist’ way. We have all, no doubt, experienced the difficulty students have in ‘abstracting’ facts from the results of practical work. (Driver and Bell, 1986: 443–4) A constructivist view of knowledge is often critiqued as being culturally shaped or comprised of culturally relative truths. In this case, knowledge is portrayed ‘as a result of activity within the scientific community as relativist and solely the result of social processes . . .’ (Driver, Asoko et al., 1994: 6). This critique of relativism has at times been made in science education (for example Matthews, 1992), implying ‘anything goes’. But a balance (Hodson, 1998: 17) between a reality of the natural world and social interactions is, however, advocated in a realist ontology.
Constructivist views of learning
37
But, a view of scientific knowledge as socially constructed does not logically imply relativism. In proposing a realist ontology, Harré (1986) suggests that scientific knowledge is constrained by how the world is and that scientific progress has an empirical basis, even though it is socially constructed and validated. Whether or not a relativist perspective is adopted, however, the view of scientific knowledge as socially constructed and validated has important implications for science education. It means that learning science involves being initiated into scientific ways of knowing. Scientific entities and ideas, which are constructed, validated and communicated through the cultural institutions of science, are unlikely to be discovered by individuals through their own empirical inquiry; learning science thus involves being initiated into the ideas and practices of the scientific community and making these ideas and practices meaningful at an individual level. The role of the science educator is to mediate scientific knowledge for learners, to help them to make personal sense of the ways in which knowledge claims are generated and validated, rather than to organise individual sense-making about the natural world. This perspective on pedagogy, then, differs fundamentally from an empiricist perspective. (Driver, Asoko et al., 1994: 6) Hence, Carr et al. (1994) suggested, with respect to the concepts of energy, floating and sinking, and photosynthesis, that: teachers reflecting on their pedagogy could usefully consider five questions as they develop their teaching approaches and reflect on student learning. These are: • • • • •
Does nature contain a definition of the concept which can be uncovered through appropriate experiences? How does science develop a statement of a concept? Is there a single explanation for a phenomenon which teachers should aim at? Can science always provide an answer to a question? When a ‘better’ explanation is proposed how do scientists decide to accept it? (Carr et al., 1994: 151)
However, Gilbert (1997), in her postmodernist study of science education, argued: Thus within the deconstructive approach to science education, the knowledge claims of science are assessed, not with respect to the a priori ‘reality’ of nature assumed in modernist thought, but with respect to the ‘rules’ on which the discourse of science is founded, with respect to the actions and
38
Constructivist views of learning
practices of scientists. These rules, actions and practices are ‘real’, in the same way that nature, for modernist thinkers, is the reality with respect to which all knowledge claims must be checked. Consequently, it is not the case, in post-modernist epistemologies, that all knowledge claims are equally valid, or that their status is ‘relative’; the truth status of a particular claim is easily decided, once the rules of the discourse within which they have been made are known. (Gilbert, 1997: 365–6) Here ‘rules’ is used in the sense of Wittgensteinian language game; that is: As having certain highly specific and important ‘rules’ which, in order for the ‘game’ to take place, must be ‘known’ by the participants, then it is possible to decide, given the rules of the particular discursive field or language game, whether one knowledge claim is more ‘true’ than another . . . within any discourse, there are, very definitely, rules for the adjudication of any knowledge claims that are made, some of which are stated quite explicitly, while others exist as part of the assumed, but implicit and necessary, ‘expert’ knowledge of the participants. (Gilbert, 1997: 365) In teaching science, consideration is also given to the notions of both personally and socially constructed knowledge (Driver, Asoko et al., 1994). On the one hand what the learners are doing during active learning may be seen as the personal construction of knowledge – each learner must construct the knowledge for him or herself, since the teacher cannot do it for him or her. On the other hand, the learner is constructing knowledge that is part of the socially constructed and consensually agreed knowledge of the community of scientists. As Driver states: Learning science, therefore, is seen to involve more than the individual making sense of his or her personal experiences but also being initiated into the ‘ways of seeing’ which have been established and found to be fruitful by the scientific community. Such ‘ways of seeing’ cannot be ‘discovered’ by the learner – and if a learner happens upon the consensual viewpoint of the scientific community he or she would be unaware of the status of the idea. (Driver, 1989: 482) Hence, later constructivist consideration of the notions of both personal and public knowledge is linked to the distinction between students learning public science and scientists constructing new public knowledge. This raises a concern for many teachers of science. What ‘weight’ should be given to children’s ideas compared to the accepted scientific view (Millar, 1989)? When students have been engaged in teaching and learning activities, have developed their ideas, but have not acquired the scientific concept, what do
Constructivist views of learning
39
you do? Is it acceptable to say the student has developed his or her personal constructions and therefore the teaching has been successful? Or is the hallmark of effective teaching whether the students have acquired the scientific view? When does the teacher make the professional decision that this student has learnt as much as he or she is able at this point? Millar (1989) acknowledged the ‘pedagogical problem of teaching science as an agreed body of knowledge, as public rather than personal knowledge’ (p. 592) when the learning process is one of reconstruction of personal knowledge. Further aspects of the 1990s critiques (and responses) on constructivism, namely theory adjudication, science curricula, progression in the curriculum, the capabilities and learning styles of students, the social dimension to the practice of science and the learning of science, the role of the teacher, the scope of the utility of constructivism, the status of constructivism with teachers, the impact on teacher development and links to progressive education, are given in Bell and Gilbert, 1996: 52–7.
40 Social views Chapter 3
of learning
Social views of learning
Social constructivism One of the main criticisms of personal constructivism was that it ignored the socially and historically situated nature of knowing. It gave ‘primacy to abstract mental structures and rational thought processes at the expense of the historically and socially constituted subjectivity that learners bring to the reasoning process’ (O’Loughlin, 1992: 800). In response, there was a growing recognition of the role of the social and cultural aspects in learning in science as well as the personal, constructivist aspects, and science educators sought to develop a social constructivist view of learning (Driver, Asoko, Leach, Mortimer, and Scott, 1994; Solomon, 1987; Tobin, 1990) which paralleled the recognition given in the literature on human development (Olssen, 1991) and learning (Nuthall and AltonLee, 1993; Resnick, 1991). However, social constructivism has early and multiple origins. For example, Berger and Luckmann (1966) argued that what passes for ‘knowledge’ in society is not just the theoretical knowledge of the kind that academics might concern themselves with, but also commonsense knowledge – that which guides people in everyday life, through routines, habits, and patterned behaviour. They stated that ‘society is a human product . . . Man (sic) is a social product’ (p. 79). Hence, Berger and Luckmann acknowledged both the personal and the social. The social constructivist position with respect to education is of long standing. Schutz and Luckmann (1973) pointed out that the extensive socially mediated learning of young people, derived from interaction with their parents and peers as well as from watching television, can lead to a well-developed ‘life-world’ knowledge in any field which might conflict with what schools are trying to teach about the same field. This split between life-world knowledge and school knowledge is well documented in the field of science (see, for example, Bell, 1981c; Solomon, 1983). The term ‘social constructivism’ was not well defined in the science education literature and was usually used to make a contrast with personal constructivism and to acknowledge the sociocultural aspects of learning. The definition given by Driver, Asoko, Leach, Mortimer and Scott (1994) was that ‘a social constructivist
Social views of learning
41
perspective recognises that learning involves being introduced to a symbolic world’ (p. 5). This definition implied that cognition is not bounded by the individual brain. Cognition is seen as involving the mind, a social process, and not just cognition about social processes. The social and personal are intertwined, as the social context in which a cognitive activity takes place is an integral part of the activity, not simply its context (Resnick, 1991; Salomon and Perkins, 1998). Socially constructed knowledge is both the medium for and the outcome of human social interaction. By the mid-1990s, a criticism of social constructivist views of learning in science education was the underdeveloped and vague use of terms and concepts, for example ‘socially constructed knowledge’. This lack of definition gave rise to questions such as: where is the socially constructed knowledge? What is/are the process(es) by which social beliefs and knowledge are constructed? With what criteria are socially constructed beliefs and knowledge in science education judged and evaluated by the group and by individuals? In what ways is the social construction of knowledge linked with rational and other ways of knowing, for example, intuition? What is the role of the affective aspects of learning in the social construction of knowledge? Are these fruitful questions to ask? As part of the Learning in Science Project (Teacher Development), Bell and Gilbert (1996) argued that there are two important criteria with which to review these different social views of learning. The first is the extent to which a view considers not only the culture of the classroom, but also the wider sociocultural views of society. The second is the extent to which they give consideration to the reconstruction of the social as an individual interacts with it. Most accounts view the individual as changing in response to the social. For example, the terms ‘enculturation’, ‘socialisation’, ‘introduction to the culture’, ‘appropriate the cultural tools’, and ‘arriving on a foreign shore’ emphasise that the individual’s personal constructions develop during the learning process towards the socially shared and agreed to knowledge. A social constructivist view of learning was proposed by Bell and Gilbert (1996), who supported a view of learning in teacher development which considers both the development of the individual’s construction of meaning towards the socially agreed to knowledge and the reconstruction and transformation of the culture and social knowledge itself. In other words, such a view of learning would acknowledge the partially determining and partially determined characteristic of human agency – the interaction of the individual with the social can change both. The personal construction of knowledge is mediated by socially constructed knowledge and the social construction of knowledge is mediated by personally constructed knowledge. Such a view could position teachers as agents, empowered and legitimate speakers, constrained by the social. The issue of power in social discourse was addressed in poststructuralist views of knowledge, knowing and development (Davies, 1993; Lemke, 1990). Similarly, Cobb (1994) stated that ‘mathematical learning should be viewed as both a process of active individual construction and a process of enculturation
42
Social views of learning
into the mathematical practices of wider society’ (p. 13) – the description could also be applied to learning science and to learning by teachers. Cobb viewed the two perspectives – constructivism and the sociocultural – as each telling half the story. Each perspective implies the other but foregrounds one aspect only. Bell and Gilbert (1996) supported a view of learning (with respect to teacher development) that considered both the development of the individual’s construction of meaning towards the socially agreed to knowledge and the reconstruction and transformation of the culture and social knowledge itself. In other words, such a view of learning would acknowledge the partially determining and partially determined characteristic of human agency – the interaction of the individual with the social can change both. The personal construction of knowledge was seen as mediated by socially constructed knowledge and the social construction of knowledge was seen as mediated by personally constructed knowledge. In summary, Bell and Gilbert (1996) included the following points in a social constructivist view of learning with respect to teacher development: •
• • •
• •
•
•
Knowledge is constructed by people. This is termed the trivial component of constructivism by von Glasersfeld but, as Solomon (1994) points out, it is an ill-founded description. It is the aspect of a constructivist view of learning that is referred to most widely in science education. Construction and reconstruction of knowledge is both personal and social. Learning involves the interaction of the personal and the social construction of meanings, and both may be changed in the interaction. Socially constructed knowledge is both the context for and the outcome of human social interaction. The socially constructed knowledge is an integral part of the learning activity. Learners as developing people have partial agency. They are partially determining and partially determined. Social interaction – for example, in dialogues, accounting and narratives – promotes learning of socially constructed knowledge, personal construction of meaning, and the reconstruction of social knowledge. Learners can reconstruct their knowledge through reflection. Metacognition is an important part of learning and can involve reflection on the degree of understanding or the nature of the thoughts. Knowledge is not something in the world to be discovered (as in a discovery learning approach). We do not learn by ‘reading the book of nature’. Rather, people construct mental representations of phenomena and these mental constructions are constrained by how the world is (Carr et al., 1994; Driver Asoko et al., 1994). In particular, scientists conduct experiments in order to test the degree of fit between their constructions and how the world seems to be. Both a realist and a relativist position can accept that we can never directly know the ‘real world’ or an ‘absolute’ truth. A suitable approach to the issue of realism versus relativism might be to follow Kelly (1969), that is,
Social views of learning
•
43
to put the matter to one side; or von Glasersfeld (1991), that is, to reject the concept of reality; or to see the debate as still in progress. All the approaches to constructivism take a similar view on communication: a directly transmitted and received message is not possible; there exists only an active construction of meaning by the hearer.
A social constructivist view of learning was used in the theorising of the Waikato theses of Mavis Haigh (1998) and Linda Wilkinson (1993).
Sociocultural views of learning By the mid to end of the 1990s, in response to the critiques of constructivism – both within and from outside – a wide variety of social views of learning were used in the Waikato research, for example in the Learning in Science Project (Assessment), in which learning was theorised as a sociocultural activity (Bell, 2000; Bell and Cowie, 2001b). (A review of this work, adapted from Bell and Cowie, 2001b, is given here.) In the late 1990s, debates within the science education, education and other literature on learning were clustered around sociocultural views of cognition and learning which are variously described as social cognition (Augoustinos and Walker, 1995; Resnick, 1991; Salomon and Perkins, 1998); social constructivist view of learning (Bell and Gilbert, 1996; Driver, Asoko et al., 1994); situated learning (Hennessy, 1993; Lave and Wenger, 1991); apprenticeship, guided participation, participatory appropriation (Rogoff, 1995); distributed cognition (Carr, 1997; Salomon, 1993a, b); mediated action (Vygotsky, 1978; Wertsch, 1991; Wertsch, Del Río and Alvarez, 1995) and discursive activities (Bell, 2000; Gilbert, 1997; Harré and Gillett, 1994). These categories and associated descriptions are not mutually excluding; there is much overlap and lack of clarity. This is due to various categories having been developed from within different disciplines, for example, anthropology, sociology, psychology, and education. Different and similar words are used, different meanings are constructed and different emphases highlighted. Learning (teaching and formative assessment) were theorised as purposeful, intentional activities involving meaning making; an integral part of teaching and learning; a situated and contextualised activity; a partnership between teacher and students; and involving the use of language to communicate meaning. Since the mid-1990s, there has been considerable theorising on learning with respect to sociocultural views of learning in education and in science education, with the goal of considering both individual and social aspects in the process of meaning making. In other words, the mental activities of individuals and an individual’s meaning making are considered along with their socially and historically situatedness. Salomon and Perkins (1998), in arguing the case for something called ‘social learning’, distinguished six meanings of social learning for the sake of conceptual clarity. The first three are of interest here:
44
Social views of learning
1
Socially mediated individual learning. For example, a teacher (the mediating agent) ‘teaches’ a student (the individual learner), the concept of ‘black holes’ through one-on-one tutoring. This approach views the social system (the teacher–student discussion and use of language) as enhancing the individual’s learning as an individual, striving to improve mastery of knowledge and skill. Social mediation as participatory knowledge construction, that is, learning is the participation in a social process of knowledge construction. For example, the students co-construct the already known scientific knowledge of ‘volcanoes’ for themselves, during small group discussion activities such as peer tutoring, collaborative, cooperative and reciprocal learning. This approach views the learning process of both a social and individual process – the two cannot be separated out. Social mediation by cultural scaffolding. That is, the learner is helped in some way by cultural artefacts, for example tools such as computers, and sign systems such as speech genres. For instance, the student may learn about the concept of electric current through the learning and use of mathematical language in the equation: V=IR. The teacher may demonstrate the use of the equation on the whiteboard, followed by the teacher doing it with the student at his or her desk, followed by the student doing it on her or his own, without the teacher support or scaffolding. This approach views the learning as the appropriate use of language as a cultural tool after Vygotsky (1978).
2
3
These three meanings of ‘social learning’ underpin sociocultural views of learning in which the learning includes not just cognitive practices but also sociocultural practices. Here, the sociocultural views of learning – situated learning, distributed cognition, and mediated action – will be detailed with respect to science education. Learning (teaching and assessment) as a situated activity One sociocultural view of learning is that of learning as a situated activity. ‘Situated activity’ is a phrase used by Lave and Wenger (1991) to locate learning in the processes of social interaction, not in the heads of individuals. In other words, learning is seen as a process that takes place in a co-participation or co-constructivist framework, not in an individual mind. Lave and Wenger (1991) made a case for focusing on the relationship between learning and the social situations in which it occurs. Rather than defining learning in terms of acquisition or internalisation of structure, they viewed learning as the increased access of learners to participating roles in expert performances. Learners are seen as ‘participating in communities of practitioners and that mastery of knowledge and skill requires newcomers to move toward full participation in the sociocultural practices of a community’ (Lave and Wenger, 1991: 29). Hence, learning is
Social views of learning
45
seen as an integral and inseparable aspect of social practice or, in other words, the process by which newcomers become part of a community of practice. Lave and Wenger (1991) used the term ‘legitimate peripheral participation’ to denote this learning through apprenticeship: ‘legitimate peripheral participation is proposed as a descriptor of engagement in social practice that entails learning as an integral constituent’ (p. 35). The novice comes to think and perceive as well as behave like the expert (Nuthall, 1997), in a process labelled ‘appropriation’ (Rogoff, 1993). Nuthall (1997) distinguished between the ‘appropriation’ of an expert’s knowledge and skills and the concept of ‘internalisation’ that cognitivist theorists use to describe the acquisition of mental skills: Whereas internalisation refers to the incorporation of behaviour and knowledge into the cognitive processes of an individual mind, appropriation is the process by which two people come to understand each other and work effectively together. They each appropriate the product of their mutually evolving partnership in the activity. The process is inherently mutual, creative and situation specific (Rogoff, 1993). (Nuthall, 1997: 705) The term ‘enculturation’ is also used to describe the process: . . . learning is a process of enculturation or individual participation in socially organised practices, through which specialised local knowledge, rituals, practices, and vocabulary are developed. The foundation of actions in local interactions with the environment is . . . the essential resource that makes knowledge possible and actions meaningful. (Hennessy, 1993: 2) In other words, social processes can be seen as an integral part of cognition (Resnick, 1991). Enculturation underpinned the theorising used by Meiying Chu in her doctoral work (Chu, 1997). She interviewed 28 Asian students, aged 14–20, about their perceived different learning and teaching approaches, different ways of understanding what learning is, different expectations of what it means to be a student, and the differing roles of the teachers in the education systems of New Zealand and their original countries of Taiwan, Hong Kong or People’s Republic of China. She argued that the traditional Chinese culture influenced the learning of the Asian–Chinese students in the New Zealand context, and proposed an enrichment model of learning, rather than earlier deficit models, in which Asian students were perceived as lacking attributes, when compared to local students. The enrichment model of learning incorporated the contributions of the traditional Chinese culture of the learners as well as western culture, and proposed that this double perspective on learning and teaching develops metacognition and encourages better learning.
46
Social views of learning
Enculturation and situated learning was also used to theorise the learning of some Waikato tertiary science students during work placements in scientific research institutions and scientific businesses, as a part of a cooperative education programme (Eames, 2003a). The tertiary science students were enculturated into the social practices of the community of scientists. Enculturation may also include the development of dispositions as in the Waikato doctoral research into learning technology in an early childhood situation (Carr, 1997) and the learning by secondary students of critical thinking dispositions, such as to seek clarification, in the context of science lessons (Hameed, 1997). New theories of learning, such as social constructivism, argue that ability-based instructional efforts are too narrow because human behaviour is determined by both abilities and tendencies such as inclinations and sensitivities. A recent conceptualisation of a theory of thinking which includes these components is the Dispositional Theory of Thinking (Perkins, Jay and Tishman, 1993; Tishman, Jay and Perkins, 1993; Tishman, Perkins and Jay, 1995). This theory postulates that seven dispositions, each comprising abilities, inclinations and sensitivities specific to the disposition, are necessary for quality thinking. These are: • • • • • • •
to to to to to to to
be broad and adventurous; sustain intellectual curiosity; clarify and seek understanding; be planful and strategic; be intellectually careful; seek and evaluate reasons; be metacognitive.
Tishman, Jay and Perkins (1993) contend that dispositions are acquired through social interactions, and that enhancing thinking can be fruitfully viewed as enculturation into a community of practice. The student is immersed in the culture, exposed to social practices (language, models and interactions) that emphasise thinking. However, while some researchers talk of enculturating and socialisation of students into the norms and practices of scientists, what is required, according to Jane Gilbert (1997), is this plus developing students’ ability to use or refuse (Fuss, 1989) the discursive practices of science. Learning (teaching and assessment) as distributed cognition Another sociocultural view of learning is that of learning as a distributed cognition. Situating cognition in social practice leads to a view of cognition as distributed across the context in which it occurs – hence the term ‘distributed cognition’. When studied in real life situations (for example, in planning the family holiday), people appear to think in conjunction or partnership with others and with the help of cultural tools and artefacts (for example language, maps and
Social views of learning
47
computers). A distributed view of cognition has been largely developed by those interested in the use of technology (for example computers) in learning. A strong version of distributed cognition is that ‘while individuals’ cognitions are not to be dismissed, cognition in general should be re-examined and conceived as principally distributed’ (Salomon, 1993b: xv, italics in original). The weaker version is that solo and distributed cognition are still able to be distinguished from each other and are taken to be in a dynamic relationship. It is this weaker version that is primarily discussed in this section, for Salomon (1993a: 113, italics in original) reminds us that ‘not all cognitions, regardless of their inherent nature, are distributed all the time, by all individuals regardless of situation, purpose, proclivity or affordance’. Also, different writers in the field of distributed cognition have differing views on the degree of distribution of cognition. Hence, thinking can be considered to involve not just ‘solo’ cognitive activities but also distributed ones – distributed across other people and the sociocultural situation. Cognition is not seen as merely in-the-head activities, decontextualised tools and products of the mind (Salomon, 1993a). Nor is cognition seen as residing in the heads of individuals, with the social, cultural and technological factors relegated to the background. Distributed cognition can be summarised as referring to the following: 1
2
The surround – the immediate physical and social resources outside the person – participates in cognition, not just as a source of input and a receiver of output, but as a vehicle of thought. The residue left by thinking – what is learned – lingers not just in the mind of the learner, but in the arrangement of the surround as well . . . (Perkins, 1993: 90)
The social and artefactual surrounds, alleged to be ‘outside’ the individuals’ heads, are not only sources of stimulation and guidance but are actually ‘vehicles of thought’. Distributed cognitions do not have a single locus ‘inside’ the individual. Rather, they are said to be ‘in between’ and are jointly composed in a system that comprises an individual and peers, teachers or culturally provided tools. ‘Distributed’ is used in the sense of ‘stretched over’ (Salomon, 1993a) rather than just divided up. While not all cases of distributed cognition can be viewed as the same, they are seen as having one important quality: ‘the product of the intellectual partnership that results from the distribution of cognitions across individuals or between individuals and cultural artifacts is a joint one: it cannot be attributed solely to one or another partner’ (Salomon, 1993a: 112). The environment in the science classroom or school laboratory provides social, physical and artefactual support for cognition in science. Artefacts that help students think may be tools such as calculators, computers; symbolic representations such as language, mathematical symbols, graphs, diagrams; the physical environment, such as laboratory benches; and the social support of discussions with the teacher and other students (Pea, 1993). Human cognition can be seen as
48
Social views of learning
distributed ‘beyond the compass of the organism proper in several ways: by involving other persons, relying on symbolic media, and exploiting the environment and artifacts’ (Perkins, 1993: 89). The social, artefactual and physical support in the surrounds can enable a person to deal with complex concepts that would be unmanageable for one person. This can be exemplified in the case of two students working together in front of a computer screen, discussing the effects on the health of a person of changing the variables in a computer sumulation model of human nutrition. Cognition is also shaped by the situation with respect to affordances (Pea, 1993). Some technology affords greater opportunity for higher order kinds of thinking and learning (Carr, 1997; Perkins, 1993). ‘Affordance’ refers to ‘the perceived and actual properties of a thing, primarily those functional properties that determine just how the thing could possibly be used. Less technically, a doorknob is for turning, a wagon handle is for pulling’ (Pea, 1993: 51). In the educational setting, we hope that we can get a learner to attend to the relevant properties of the environment or object or text, such that the learner can join in. There will be variation in the ease with which a social, cultural, technological or environmental tool can be conveyed to and used by a learner in activities which contribute to distributed cognition. In summary, distributed cognition is one way of viewing socially constructed learning and knowledge – knowledge that is ‘socially constructed, through collaborative efforts toward shared objectives or by dialogues and challenges brought about by differences in persons’ perspectives’ (Pea, 1993: 48). Learning (teaching and assessment) as a mediated action A third sociocultural view of learning and cognition is that they are mediated actions (Vygotsky, 1978; Wertsch, 1991). A mediated action is a human action that employs mediational means, such as technical tools – for example a computer, and psychological tools – for example signs, such as languages. As with situated and distributed views of cognition and learning, the focus is on human action in context. Again, the basic goal of these (and the other) sociocultural approaches to the mind is to create an account of human mental processes that recognise the essential relationships between these mental processes and their social, cultural and institutional settings (Wertsch, 1991). For example, a teacher and student co-construct an understanding of the learning required to reach the desired level of attainment, during formative assessment involving both feedback and feedforward (Cowie, 2000). A fundamental assumption of this sociocultural approach to the mind is that the unit of description and analysis is ‘human action’. To understand mental functioning, one cannot begin with the environment (as in a behaviourist approach) or a human agent in isolation from the sociocultural settings (as in the cognitivist approach). Instead, a sociocultural view assumes that the notion of mental functioning is not limited to processes of the brain of the individual, and
Social views of learning
49
that it can be applied to social as well as individual forms of activity. In a sociocultural view of the mind, what is discussed and explained is human action and interaction. Wertsch (1991) asserts that in studying human action, one sees a close relationship between social communicative processes and individual psychological processes. Hence, to understand the individual, it is necessary to understand the social relations in which the individual exists. ‘Mediated action’ is a term used by Wertsch (1991) to emphasise that human action typically employs ‘mediational means’ such as tools and signs (including language). He gives support to Vygotsky’s (1978) claim that the higher mental functioning and human action in general are mediated by tools (or technical tools) and signs (or psychological tools). For example, a science student uses the psychological tools of language, counting systems, mnemonic techniques, algebraic systems, writing, diagrams, maps to learn and understand science, and these are a part of the tool kit available to humans in the meaning-making process. A defining property of higher mental functioning is the fact that it is mediated by tools and by sign systems such as natural language. The incorporation of the mediational means does not simply facilitate action that could have occurred without them (Wertsch, 1991); instead, as Vygotsky (1978) noted, by being included in the process of behaviour, the psychological tool alters the entire flow and structure of mental functions. Hence the agent and the means become inseparable. ‘The action and the mediational means are mutually determining’ (Wertsch, 1991: 119). For example, Wertsch (1991) gives the example of the blind person’s stick. The stick is a particular shape and colour due to its use by a blind person. One cannot separate the stick and the blind person to make sense of it. In a similar way, the bunsen burner used in school laboratories is a particular shape and size due to its use by school students. The burner differs from that in a kitchen or in an industrial process. In summary, mediated action rests on assumptions about the close relationship between social communicative processes and individual psychological processes. The processes and structures of mediation provide a crucial link between historical, cultural and institutional contexts and mental functioning. It is the sociocultural situatedness of mediated action that provides this essential link between the cultural, historical and institutional setting on one hand and the mental functioning of the individual on the other. In science education, the social and psychological processes emphasised are those of scientists. Summary The above sociocultural views of learning and thinking – situated, distributed and mediated action – all have in common aspects that are useful in theorising about learning (and teaching and assessment) in science. These are now summarised. The main goal of a sociocultural view of learning, thinking and the mind is to create an account of human mental processes that recognise the essential relationships between mental processes and their social, cultural and institutional
50
Social views of learning
settings (Wertsch, 1991). In terms of the science classroom, the goal is to account for the way social practices, including language, determine how and what children think and learn. Sociocultural views of learning inform us that it is the whole of what goes on in classrooms that determines the learning, not just what is happening inside an individual’s head. Overall, the sociocultural perspective was developed from a desire to see school learning within a larger cultural context. This led to a focus on the culturally embedded nature of the classroom processes and the central role that cultural norms and artefacts play in structuring the learning and the way we view learning. (Nuthall, 1997: 711) Likewise, learning (as well as teaching and assessment) is a highly contextualised and situated activity. In understanding learning in science, we need to consider not just the meaning making by an individual, but the context in which it is occurring (Bell and Cowie, 2001b). Secondly, meaning is central to a sociocultural approach to mind. A sociocultural view emphasises the ‘mind’ rather than the ‘brain’. If thinking is viewed as situated, distributed or mediated action, then the mind is more than cognition or brain processing. It includes a wide range of psychological phenomena, such as mental processes, self, emotions, intentions. Mind, in this view, goes beyond the skin and so we call it ‘socially distributed’, as mind and mediated action cannot be tied to an individual acting in vacuo. Mind, as it is used by Wertsch (1991), is defined in terms of its inherently social and mediational properties. In science lessons, we as teachers interact with the mind, not the brain of students. Thirdly, sociocultural approaches consider both the individual and the social aspects of learning and thinking, given that the goal of a sociocultural approach to learning is to ‘explicate the relationships between human mental functioning, on one hand, and cultural, institutional and historical situations in which this functioning occurs, on the other’ (Wertsch, del Río and Alvarez, 1995: 3). There is a need for such a sociocultural view as previous views of learning saw the learner as internalising knowledge, whether ‘discovered’, ‘transmitted’, ‘experienced in interaction’ or ‘constructed’ (Lave and Wenger, 1991). These previous views established a dichotomy between inside and outside, and between the individual and the social, especially in individualistic, reductionism psychological debates. In contrast, sociocultural views focus on the ‘mind’ (rather than just the ‘brain’); human action (rather than behaviour); and meaning making (rather than linguistic structure or mental/conceptual representation) (Bruner, 1990; Wertsch, 1991; Wertsch, del Río and Alvarez, 1995). Sociocultural views of learning, then, specifically address the issue of the distinction between the ‘individual’ and the ‘social’ in past psychological debates. Learning is seen as involving both individual construction of meaning and the social aspects of the school and scientific communities.
Social views of learning
51
For example, Cobb (1994) asserted that ‘mathematical learning should be viewed as both a process of active individual construction and a process of enculturation into the mathematical practices of wider society’ (p. 13) – the description could also be applied to learning science. Cobb views the two perspectives – constructivism and the sociocultural – as each telling half the story. Each perspective implies the other but foregrounds one aspect only. Salomon and Perkins (1998) also see the need to consider both, stating that one cannot be reduced to the other. Rogoff (1995) addressed the social and individual issue by proposing a sociocultural approach ‘that involves observation of development in three planes of analysis corresponding to personal, interpersonal, and community processes’ (p. 139). These are described as ‘inseparable, mutually constituting planes comprising activities that can become the focus of analysis at different times but with the others necessarily remaining in the background of the analysis’ ( p. 139). One can become foregrounded, whilst the other two are not ignored, but backgrounded. She asserted that the development of children (for example) occurs through a process of participation and collaboration in social activities. These social activities can be in personal, interpersonal and community processes. The use of activities as the unit of analysis enables social, individual and cultural environments to be described in relation to each other, for none is seen to exist separately (Rogoff, 1995). The activities on which she focuses in each plane are: apprenticeship in the community plane; guided participation in the interpersonal plane; and participatory appropriation in the individual plane. Hence, sociocultural perspectives on human functioning emphasise the relationship between mental processes and the sociocultural setting. Salomon (1993a), in the debate on the relationship between and the relative roles of the individual and distributed cognitions, proposed a model for the interaction between individual and distributed cognitions. He described the components as interacting with each other in a ‘spiral-like fashion, whereby individuals’ inputs, through their collaborative activities, affect the nature of the joint, distributed system, which in turn affects their cognitions such that their subsequent participation is altered, resulting in altered joint performances and products’ (Salomon, 1993a: 122). This spiral-like development allows for distributed cognitions and one’s own ‘solo’ competencies to be reciprocally developed. Hence, the relationship will develop over time. This position of the sociocultural views of learning, that are accepting of both social and individual learning and that differentiate between thinking and language, is appealing to theorising on, say, formative assessment in science education because the teachers and learners do attend to the social aspects of learning in the classroom, even though the education system as a whole (and in particular, assessment) focuses on the individual. This view of individual and social aspects of learning is also addressed by Hodson (1998) in his discussion of exploring and developing personal, critical understanding of science to ‘equip students with the capacity and commitment to take appropriate, responsible and effective action on matters on social,
52
Social views of learning
economic, environmental, and moral-ethical concern’ (p. 21) – personalised science. The fourth aspect in common between the three sociocultural views of learning is that of the methodological concern of the unit of analysis. One way to study both aspects (of individual and social aspects of learning) is to adopt the unit of analysis of human action (Wertsch, 1991), rather than focusing on the unit of analysis of concepts, linguistic and knowledge structures, attitudes, as often found in psychology, although they might be used in an analysis of human action. An example of human action within the context of science education would be the action or practice of a teacher helping a student to wire an electric circuit correctly in parallel to another. The thinking of the teacher, the use of language and technology by the teacher, and the teacher’s intentions cannot be separated from the action of manipulating the wires in a specific way. All aspects need to be addressed if meaning is to be made of the situation. In his analysis, Wertsch (1991) sees mediated action as the irreducible unit of analysis and the person-acting-with-mediational-means as the irreducible agent involved. In a similar way, distributed cognition recognises that some activities are so highly contextualised, and dependent on the situation, that we cannot easily make the distinction between cognitive knowledge and skills, the context, and the activity a person is engaged in. In effect, the unit of analysis for research and theorising on learning has changed from the individual alone to the individual plus those parts of the surround that may be supporting the cognition. Or, as Perkins (1993) described it, the unit of analysis has changed from the ‘person-solo’ to the ‘person-plus (the surroundings)’. Likewise, Pea (1993) takes the ‘person-in-action’ as the unit of analysis. That is, the unit of analysis is the person plus the ‘resources that shape and enable activity [that] are distributed in configuration across people, environments and situations’ (Pea, 1993: 50). In other words, cognition emerges or is accomplished, rather than being possessed. In using the ‘person-in-action’ as the unit of analysis, we need to consider the role of intent, desire and conation which shapes both their interpretation and use of resources for the activity. In this way, sociocultural views of learning address the integration of cognition, affect and conation, in a way that constructivist approaches do not. Sociocultural views of learning have been used as the theorising in Waikato research theses, for example Chang, 2000; Cheng, 2000; Cowie, 2000; Eames, 2003a; Hameed, 1997, and the international literature, for example Brickhouse, 2001; Lemke, 2001; Roth, 1995, 1998.
Learning as a discursive practice Another way of viewing learning (teaching and assessment) in science is to give consideration to the central role played by language in meaning making; the partnership between the teacher and students, and communication. The role of language is theorised as a discursive practice. Discursive views of learning, in
Social views of learning
53
contrast to the other three sociocultural views already discussed, theorise on only the social aspects of learning and give no functional value to a consideration of the individual aspects (Bell and Cowie, 2001b). If we view learning (and formative assessment) as discursive activities, we are predominantly giving attention to language-in-use. In the science classroom, we are giving focus to the ways language, and scientific language in particular, is used to promote thinking and learning of science. The role of language (and other symbol use) is central to discursive psychology, in which we are examining human functioning in actual social and cultural settings. A discursive activity or practice is: . . . the repeated and orderly use of some sign system, where these uses are intentional, that is, directed at or to something . . . Discursive activities are always subject to standards of correctness and incorrectness. These standards can be expressed in terms of rules. Therefore, a discursive practice is the use of a sign system, for which there are norms of right and wrong use, and the signs concern or are directed at various things. (Harré and Gillett, 1994: 28–9) In short, a discursive activity is an intentional, normative action, using sign systems. The focus of discursive psychology is what talk and writing is being used to do, that is, what language is being used to achieve, rather than language been seen as an abstract tool to state or describe things. Language is seen as functional and used by people, for example to justify, explain, blame, excuse, persuade and present an argument. Hence, the notion of language-in-use relates to that of communication. In the science classroom, we are giving focus to the way the language of science is used to construct descriptions, explanations and arguments and to validate knowledge claims. We are ‘talking science’ (Lemke, 1990). As with the other sociocultural views of learning, meaning is central when considering learning (teaching and assessment) as discursive practices. By knowing what a situation means to a person, we are able to understand what that person is doing, for example when an earth scientist is collecting rock samples as data (Harré and Gillett, 1994). We understand the behaviour of an individual when we grasp the meanings that are informing a person’s activity: [ Wittgenstein] came to realise that understanding and the phenomena of meaning or intentionality in general could only be approached by looking at what people actually do with word patterns and other word signs. He formulated the doctrine that meaning is the use to which we put our signs. He studied the use of language in ‘language games’, by which he meant complex activities involving both the use of language and the use of physical tools and actions, where they are ordinarily encountered . . . [he] came to see that mental activity is not essentially a Cartesian or inner set of processes but a range of moves or techniques defined against a background of human activity and governed by informal rules. These rules, unlike the rules–laws at
54
Social views of learning
work in supposed inner, cognitive processes, were the rules that people actually followed. They are most evident when we consider the correct and incorrect ways of using words . . . This understanding of human activity requires us to interpret the behaviour of another according to some appreciation of the self-positioning of the subject within the complex structure of rules and practices, within which that individual moves. (Harré and Gillett, 1994: 19–20) For example, we understand the behaviour of an individual (such as a student looking under bushes on the school grounds) when we grasp the meanings (collecting samples of insects to categorise) that are informing a person’s activity. Language and other semiotic (sign) systems play an important part in producing meaning, especially meaning as it shapes human action (Wertsch, 1991). Meaning here is viewed as being produced only in a social setting, and as a process, not a fixed entity inherent in a linguistic package: Wittgenstein emphasised the interactive and conventional nature of language. As a social practice, language has no fixed meaning outside the context in which it is used. Our perception of the world is shaped by the language we use to describe it: objects, activities and categories derive their epistemological status from the definitions we create for them. Within this view, thought and language are no longer separated. ‘When we think in language, there are not “meanings” going through our mind in addition to verbal expressions. The language itself is the vehicle of thought’ (Colier et al., 1991, p. 227). (Augoustinos and Walker, 1995: 264) It is usual to think of concepts as the basis of meaning, understanding and thinking. But concepts are expressed by words and words are located in languages. Thus, the discourses constructed jointly by persons and within sociocultural groups become an important part of the framework of interpretation and meaning. The communicability of thoughts is secured by the mutual intelligibility of a shared symbolic system, such as a common language (Harré and Gillett, 1994). Scientists, science teachers and science students share a language, for example scientific language, such as sodium, Na; the universal signs for radioactive hazards; and symbols such as the mathematical symbols >, =. Being able to use these languages, signs and symbols is a discursive skill. The grasp of the use of a word/concept, such as ‘energy’, is seen as an active discursive skill, rather than an inner cognitive skill, and learning is seen as the increasingly skilled use of social practices. Therefore, the discursive view differs from the other sociocultural views in its non-mentalistic view of ‘cognition’ and ‘mind’. If priority is given to languages (words, symbols and signs) in defining what are psychological phenomena, then to present and understand cognition it must be done in terms of the ordinary
Social views of learning
55
languages through which we think, rather than looking for abstract representations of them (Harré and Gillett, 1994; Lemke, 1990). Discursive psychology considers thinking, not as a mental activity, but as the activity of operating signs (for example, language). Hence, discursive approaches to thinking and learning differ from the sociocultural views in that they see no distinction between thoughts and language. When we are talking science in appropriate ways, we are thinking science. Discursive approaches see problems in the assumptions that cognitive phenomena such as ‘attitudes’, ‘emotions’, ‘categories’ can be identified and located in an internal cognitive world – inside the head. Attention is given to the discourse itself and not the assumed underlying, internal, static mental states and processes. Instead, discursive psychology is more interested in how people discursively constitute psychological phenomena to do certain things. Psychological phenomena are ‘discursive actions’ which are ‘actively constructed in discourse, for rhetorical ends’ (Wetherell and Potter, 1992: 77). The social processes are the cognitive processes. For example, categorisation, as a psychological phenomenon, is ‘something we do, in talk’, in order to accomplish social actions (persuasions, blamings, denial, refutations, accusations) (Edwards, 1991: 94). Because ‘some constructions are so familiar, pervasive and common-sensical that they give an effect of realism or fact. People therefore come to regard some constructions not as versions of reality, but as direct representations of reality itself ’ (Augoustinos and Walker, 1995: 269). Any internal cognitive realm is conceptualised as a form of situated practice. There is no notion of internal representation or model to assume cognitive mediation (Augoustinos and Walker, 1995). Discursive (as post-structuralist) psychology is critical of social cognitive concepts such as representations, schemas, attitudes, categories which are hypothesised to be stable mental categories located within the mind. The position taken by Lemke (1990) and O’Loughlin (1992) is to deny the functional significance of individual mental processes (Nuthall, 1997). Their position is a relativist one, rather than realist, and as such may be unacceptable to many in the science education community. However, their denial is not a denial of the ‘reality’ of mind and cognition so much as a denial of the value of talk about mind and cognition (Nuthall, 1997). Hence, a discursive approach questions the notion of a knowable reality by emphasising the socio-historical and political nature of all knowledge claims. For example, the scientific knowledge claim of the early twentieth century that Maori came to New Zealand around 1100–1300 by waka (canoes) or rafts drifting to New Zealand on ocean currents is now disputed (Bishop and Glynn, 1999). The early twentieth century’s socio-historical and political contexts, within which these knowledge claims were made, denied that Maori had the intellectual ability and cultural knowledge required to navigate purposefully, rather than to drift. Coloured races were seen as inferior to European races, and hence must have drifted, not navigated. Celestial navigation by the early eastern Polynesian explorers is now documented and in a format for students (Te Tahuhu o te Mãtauranga, 1997).
56
Social views of learning
Such a post-structuralist view of psychology stresses that words do not have independent, objective meaning outside the social and relational context in which they are used: Language is viewed as reflexive and contextual, constructing the very nature of the objects and events as they are talked about. This emphasises the constructive nature and role of language . . . As people are engaged in conversation with others, they construct and negotiate meanings, or the very ‘reality’ which they are talking about. (Augoustinos and Walker, 1995: 266) People live in two worlds: the physical world and the symbolic world. The physical or material world is structured by causal processes. The symbolic world (the world of symbols) is organised by the norms and conventions of correct symbol use. It comes into being through intentional action. The relationship of a person to both these worlds can be understood through the idea of skilful action (Harré and Gillett, 1994), using complementary manual and discursive skills. To operate in the physical world, we use manual skills. To operate in the world of symbols, we need to be adept at using discursive skills. As Harré and Gillett explain: There could not be a world of symbols unless there was a material world. But these two realms do not reduce to each other. We cannot explain the world of symbols and how it works by reference to physical processes . . . there could not be language and discursive processes unless there were brains buzzing with electrical and chemical processes and there were vibrations in the air and marks on paper. But those vibrations and those marks and buzzings do not constitute the mind. They cannot explain the intentional character of symbol use and the normative constraints under which symbols must be used. A buzzing in the brain cannot be correct or incorrect. It can only be. (Harré and Gillett, 1994: 100) A discursive approach to learning enables all three aspects of ‘mind’ (cognition, affect and conation) to be taken into account, rather than each being studied in isolation. A discursive view of mind asserts that to understand the mind is to study social interaction, not the biological brain operation of an individual. For example, to understand the mind of two science students learning about electrolysis is to study the social interaction between two students discussing in the school laboratory why the electrodes in a copper chloride solution only give off bubbles at one electrode, not both (Tasker and Osborne, 1985). What is studied is not their cognition represented by words, but their discursive practices of purposefully using the language, signs and symbols to explain a phenomenon. Harré and Gillett (1994) state that we need to move away from a focus on the individual as a rational subject and to look at a broader framework to understand meaning and rule following.
Social views of learning
57
To add emphasis to the notion that communication, mental processes and conation are linked, Wertsch (1991) uses the notion of ‘voice’, after Bakhtin, for example (1986), meaning the speaking personality, the speaking consciousness. The notion of ‘voice’ is concerned with the wider issues of a speaking subject’s perspective, conceptual horizons, intentions and world view. It always exists in a social milieu, that is, not in isolation from other voices. Voices produce utterances – a notion used by Bakhtin to focus on the situated action of language-in-use, rather than on objects that can be derived from linguistic analytic abstractions. Bakhtin’s notion of utterance is linked with that of voice as an utterance that can only be produced by a voice. For example, the voice of the scientist is purposefully to describe, explain and theorise about the way things behave in our physical, technological and natural worlds. And the scientist does this with different languages, signs and symbols, and ways of knowledge validation than a theologian, who would use a different ‘voice’ and utterances. Considering how voices engage with one another is important to a discursive view of mind (Wertsch, 1991), for it is only when two or more voices come into contact (for example when the voice of a listener responds to the voice of a speaker) that meaning comes into existence. For example, during formative assessment in a science classroom, the teacher and students share their meaning making and respond through their actions to improve learning (Bell and Cowie, 2001b). Taking into account both voices reflects a concern for addressivity – the quality of turning to someone else. In the absence of addressivity, an utterance does not exist. Talking at someone, without consideration of the audience, rarely results in meaningful communication. A lecturer talking to a class of 300 tertiary undergraduate science students needs to consider the background of the students and their purposes for studying for meaningful communication to result. Addressivity is not inherent in the unit of language (for example, word or sentence) but in the utterance. The notion of addressivity means that ‘utterances are not indifferent to one another, and are not self-sufficient; they are aware of and mutually reflect one another’ (Bakhtin, 1986: 91 as quoted in Wertsch, 1991: 52). Therefore, utterances involve both a concern with who is doing the speaking and a concern with who is being addressed. A teacher, in giving feedback to a student about his or her learning of science, is concerned about speaking the voice of the scientist and how to phrase it for a learner of science. Utterances are inherently associated with at least two voices – the speaking voice may indicate an awareness of the addressee’s voice. Bakhtin’s concept of ‘dialogicality’, meaning more than one voice, is useful to Wertsch (1991). Human communicative and psychological processes are said to be characterised by a dialogicality of voices. That is, when a speaker produces an utterance, at least two voices are heard simultaneously. If human communication is characterised by a dialogicality of voices, then understanding is dialogic in nature. That is, to understand another’s utterance is to orientate oneself with respect to it. There are different sorts of dialogues: face-to-face, inner dialogue, parody, and social languages within a single national language. Dialogicality in the science classroom is illustrated in
58
Social views of learning
the work of Scott (1998, 1999), who analysed science classroom talk in terms of authoritative and dialogic nature of the discourse in the classroom. Authoritative functions of discourse are those that convey information and which emphasise the transmissive function of teacher talk. The dialogic function of teacher talk is that in which the teacher encourages students to put forward their ideas, to explore and debate points of view. In a classroom, both functions of discourse are realised – the discourse has functional dualism. The situation is dynamic as the discourse shifts between authoritative and dialogic functions. Scott (1999) suggests ‘that individual student learning (of science) in the classroom is enhanced through achieving some kind of balance between presenting information and allowing opportunities for exploration of ideas’, that is, a balance between the authoritative and dialogic functions of the discourse. Bakhtin (1986) also made a distinction between social languages (for example, teen speak), meaning discourses specific to a given social system at a given time, and national languages, for example English, French. Another notion used by Bakhtin was the notion of ‘speech genre’, giving examples such as military commands, everyday greetings and farewells. He saw the ‘speech genre not as a form of language but as a typical form [a type] of utterance’ (Wertsch, 1991: 61). The typical classroom talk pattern of teacher–student–teacher may be considered a genre. For example, the teacher asks questions, to which she or he already knows the answer, for a specific purpose – that of finding out what the students are thinking: Teacher: Student: Teacher:
What is the bone in the thigh called? Femur. Yes, that’s correct.
This kind of questioning and response pattern rarely occurs in non-teaching situations in everyday life. Wertsch distinguished between social languages and speech genres in that social languages relate to the different social groupings, whereas speech genres relate to ‘typical situations of speech communication’ (p. 61). Social languages and speech genres are good candidates for the tools in the tool kit of mediational means for meaning making, for it is through these that utterances take on meaning (Werstch, 1991). Therefore, Wertsch would view social languages, and especially speech genre, as a mediational means or tool for thinking and communication. Speech genres are seen to provide a crucial link between psychological processes as they currently exist and their cultural, historical and institutional settings. In classroom talking, the voices appropriated by the children can be fully interpreted only if one goes beyond the individual speakers involved. In order to interpret what it is that they have said and to identify ‘who’ it is that is doing the talking, one must look to the speech genre appropriated in speakers’ utterances. For example, to make sense of the action of the science student who reads the shop sign ‘No animals allowed inside’, and
Social views of learning
59
who then enters the shop, we need to consider who is doing the ‘talking’ in the sign and who is reading the sign. The talking is from the public health authorities, who have the responsibility for ensuring hygiene standards in food shops. They are using the word ‘animal’ in the everyday sense, not the scientific sense, meaning dogs, cats, snakes and other pets that people may not bring into the shop with them. The sign is addressed to pet owners. The sign is not intended for the pets themselves or for vermin that may enter the shop through holes in the floor. Nor is it intentionally referring to the millions of skin mites that all humans carry. To understand the actions of the science student going food shopping, we need to go beyond the individual. Social languages and speech genres (as mediational means) appear to be hierarchically used. Wertsch (1991) uses the term of ‘privileging’ to refer to ‘the fact that one mediational means, such as a social language or genre can be viewed as being more appropriate or efficacious than others in a particular social setting’ (p. 124). For example, in the science lesson, ‘curriculum science’ is privileged over ‘children’s science’. Formative assessment plays a role in giving students feedback as to acceptable social languages and speech genre. And as Nuthall (1997: 729) states, ‘classrooms are language communities that develop their own forms of language’. In summary, the mind, in a discursive approach, is seen as a social practice, rather than something to be sealed into its own individual and self-contained subjectivity. It is seen as a domain of skills and techniques that renders the world meaningful to the individual, rather than just the biological brain operation of an individual. The whole point of discursive psychology is to get away from ‘mythical’ mental activities (Harré and Gillett, 1994), the mind being considered as a non-mentalistic entity. This position is in contrast to the behaviourist tradition which views the mind as a private area, not available as a source of data. It is also in contrast to cognitive psychology which has a view of mental mechanisms and the existence of inner mental states and processes such as rule following (for example, the scripts of Shank and the grammars of Chomsky). Cognitive science assumes that aspects of cognition (sensation, perception, imagery, retention, recall, problem solving, thinking) are mental entities, that is, have substance. Discursive psychology is not interested in mental representations but in meanings. Thoughts are not seen as objects in the mind but the activity and essence of mind. Thoughts reside in the uses we make of public and private systems of signs. To be able to think is to be a skilled user of these sign systems, that is, capable of using them correctly. Whilst the usual meaning of ‘cognition’ is pertaining to thought, Harré and Gillet (1994) have found it useful to redefine cognition as pertaining to brain processes only. Hence, in their view, the study of mental processes can be seen as the study of discursive practices, rather than the study of internal brain functioning. One of the main criticisms of a discursive approach is its inability to explain retention and memory, for it does not focus on mental activity (Augoustinos and Walker, 1995). In highlighting or foregrounding the social, the individual mental
60
Social views of learning
aspects are hidden or backgrounded. While teachers and students may attend to the social situatedness of learning (teaching) in the classroom, assessment regimes in most education systems focus on individual achievement and cognitive development. However, Harré and Gillet (1994) explicate a neural network model of mental activity, which they assert accounts for both the abstract representations of structures and functioning in the brain and nervous systems, and the metaphorical presentations of the ‘grammatical’ structure and relationships of intended, goal-directed and norm-constrained human action (that is, discursive activities). In addition, Augoustinos and Walker (1995) argue it would be difficult to deny that cognition is taking place, for example, in reflection, learning, and deductive reasoning. A strength of a discursive view of learning and meaning is that it allows us, more readily, than the sociocultural views, to give consideration to the notion of ‘power’. If language is seen as a form of social practice and if meanings are seen as socially constructed, then what counts as coherent or meaningful depends very much on the power relationships, rather than on an absolute truth. Seen in this context, power is not the ‘possession’ of particular persons but is constituted in positions occupied by subjects in discourses. ‘Discourse’ is taken here to mean ‘a set of ideas embodied as structuring statements that underlie and give meaning to social practices’ (Monk et al., 1997: 302). This is important in learning, teaching and assessment tasks in science education, where the power relations between teacher and student are influential on pedagogical and learning outcomes. Foucault’s notion of surveillance and other techniques of power (Gore, 1998) might be a useful start from which to theorise about power in formative assessment as a subtle form of social control.
Summary In summary, the Waikato research on learning has developed from the early work on personal constructivism in the 1980s, through social constructivism in the early 1990s, to sociocultural views and discursive views of learning in the late 1990s, although other views of learning, for example the humanism and enactivism, have been referred to as well (Barker, 2001). No doubt these views of learning will continue to develop as the critiques of the ontology and epistemology of constructivism are developing (Fox, 2001; Gil-Pérez et al., 2002; Jenkins, 2000; Leach and Scott, 2003; Niaz et al., 2003); discussions on the dual aspects of learning – individual and social (Leach and Scott, 2003); and the implications for pedagogy (Leach and Scott, in press).
Chapter 4
Pedagogies that take into account students’ thinking
61
Pedagogies that take into account students’ thinking
Introduction In the twenty years from 1979–98, the five Learning in Science Projects and related theses have researched the teaching of science, with many of the projects and theses developing and researching new teaching approaches and their effect on science learning outcomes. The Waikato research, done in the main by researchers who were also qualified and experienced teachers, can be characterised as being underpinned by a commitment to improving the learning of science. It can also be characterised as researching teaching that takes into account students’ thinking, and which is theoretically closely linked with learning theories. Over the twenty years, the research has researched, developed and refined pedagogies that take into account students’ thinking and how we as teachers respond to help the students learn science. This chapter gives an overview of the key features of this work, in: • • • • •
Taking into account students’ alternative conceptions; The Generative Teaching Approach; The Interactive Teaching Approach; The content of science; The contexts of teaching science.
Chapter 5 will continue a review of the Waikato research on pedagogies of: • • • • •
Practical work; Inclusive pedagogies; Metacognition; Classroom discussion; Teacher education and development for new pedagogies.
Taking into account students’ alternative conceptions A key finding of the first Learning in Science Project was the description of the alternative conceptions that students brought into the classroom. The development
62
Pedagogies that take into account students’ thinking
of these ideas towards the accepted scientists’ ideas, conceptual development, became a pedagogical strand in this project and later research, and identified a previously unconsidered aspect of pedagogy, that is, students’ existing ideas (Osborne and Freyberg, 1985). A total of thirteen working papers (Osborne, Freyberg and Tasker, 1980) and subsequent theses and published articles, documented the alternative (to the accepted scientific) conceptions held by students. The recognition that students brought these alternative conceptions to their learning and held onto these ideas strongly led to theorising about what pedagogical practices might be required if students’ thinking were to be taken into account (Gilbert, Osborne and Fensham, 1982; Osborne, 1982a). The main focus of this theorising was that students’ existing ideas needed to be addressed and that these ideas interacted with the taught curriculum, possibly to lead to unexpected outcomes. Teaching approaches that take into account students’ thinking typically have these components (Osborne, 1982a): 1 2 3 4 5
6
7
Teacher preparation to understand the relevant scientists’ views and the children’s views. Students being familiar with the context of the ideas, by experiencing the phenomena to be discussed. Clarification (by the students) of their own views of the phenomena being discussed. Presentation (by the students) of their own view as part of a discussion of different views and understandings. Appreciation (by students and teachers) of the views of others to create a supportive learning environment in which the personal ideas of students are valued as worthwhile contributions to the learning experiences of the class. Changing the status of the different viewpoints, so that the students will see the scientific viewpoint as more intelligible, plausible, and useful than their own. This usually involves contrasting the students’ ideas with the desired view. Elaboration of the new ideas, by students exploring similar and new examples of the phenomena to help them fully appreciate that the new ideas are intelligible, plausible and fruitful, and can be linked with more ideas in long-term memory.
These components are variously described in Osborne, Bell and Gilbert (1983) and Hewson and Hewson (1988) and the key components can be summarised as: • • • • •
students gaining a clarification and awareness of their own and others’ ideas; students exploring their ideas and testing them out, resulting in the creation of cognitive dissonance; the construction of new conceptions and/or the restructuring of existing conceptions; the acceptance of the new ideas; the use of the new ideas in familiar and new situations.
Pedagogies that take into account students’ thinking
63
Millar (1989) cautioned that a constructivist model of learning does not logically entail a constructivist model of instruction. He concluded that the ‘process of eliciting, clarification, and construction of new ideas takes place internally, within the learner’s own head. This occurs whenever any successful learning takes place and is independent of the form of instruction’ (p. 589). However, many teachers and researchers have found it helpful in terms of promoting students’ learning to consider the teaching and learning activities and approaches arising from the work of the research in children’s science, constructivism and conceptual change. Some of the studies have investigated teaching activities, such as cognitive conflict (Nussbaum and Novick, 1981; Stavy and Berkovitz, 1980); small-group discussions (Gilbert and Pope, 1986); and establishing a non-threatening learning environment (Watts and Bentley, 1987). A range of different teaching activities consistent with the view of constructivist learning is given in Bentley and Watts (1989). This includes games and simulations, role play and drama, media and resource-based activities, talking and writing activities, and problem solving. Similar suggestions are given in Grant, Johnson and Sanders (1990). The issues identified by the early research were, first, that it is hard for students to change their existing ideas. In hindsight, the proposals of the early 1980s on how to promote conceptual change seem naive. Secondly, it is appreciated that cognitive conflict and discrepant events may not result in conceptual change. The conflict situation may be perceived as threatening to the students, particularly when the student perceives that his or her self-esteem is under attack (Claxton and Carr, 1991). The conflict may be negated when the student avoids seeing or responding to the discrepancy or conflict. Observation is influenced strongly by the existing ideas of the student and what the student attends to may not be that intended by the teacher. In addition, memories of discrepant events and cognitive conflict may be reconstructed to keep intact the existing belief, rather than change the belief in the light of observations (Gauld, 1986). Even when the conflict is acknowledged by the student, he or she may not be able to replace the prior idea with a better alternative. Thirdly, many students who have modified their existing ideas in the light of teaching in classrooms may revert back to using their prior ideas in the school setting after a few months (Cosgrove, 1989; Happs, 1984). Fourthly, teachers’ concepts may influence what conceptual change occurs, if any. Kirkwood and Carr (1988a) described the varying concepts of ‘energy’ held by chemistry, biology and physics teachers. These varying concepts will mean that different learning experiences are given to the students and that the teachers will enter into different dialogues with the students. Fifthly, the sequence of the teaching activities was debated. Of particular interest was the point at which the scientific idea is introduced to the students. While much of the research suggested introducing the scientific idea alongside the students’ own ideas in the initial eliciting and clarification exercises, Rowell
64
Pedagogies that take into account students’ thinking
and Dawson (1984), drawing on Piaget’s notion of equilibration, suggested that the students’ ideas be put aside until after an initial exploration, and the scientists’ ideas studied. Then, the students could consider which view, their prior conceptions or their newly constructed ideas, best fits the evidence. The comparison between the students’ and the scientists’ ideas is delayed until the intended replacement concept is well understood by the students. Sixthly, the studies on teaching for conceptual change highlighted the mismatches in science education between what we know of how students learn and our classroom, school and system practices in science education. This is nowhere more apparent than in the area of assessment. The main issue here is how assessment that matches a constructivist view of learning can be interfaced with national or state systems of school leaver awards. In formative assessment during the learning, the focus is not so much on whether the students have achieved the standards or criteria, but to ascertain the growth or development of the students’ conceptions. Of particular importance is the teachers’ role in helping students to develop strategies to identify what they are learning. This self-assessment is part of the meta-cognition (White and Gunstone, 1989). Summative assessment presented more of a problem to constructivist teachers, especially in terms of national or state examinations in the senior school. It was argued that many of the assessment items only assessed students to ascertain whether they had the accepted scientific concept or not. Any learning or conceptual development towards the scientific concept that occurred was not recorded or credited. And lastly, another issue that has been highlighted in the studies into promoting conceptual change is the kind of science we teach in classrooms. Gunstone (1990) and Duckworth (1991) argue for keeping the science in science lessons complex. By trying to simplify it, we often make it harder to learn. By keeping the science complex, students can investigate and explore the world, their thinking and their own ideas more fully. ‘Minds-on’ science education (Duckworth et al., 1990) can then occur, not just ‘hands-on’ science. The work at Waikato to develop and evaluate new teaching approaches based on a constructivist view of learning is now reported. The third and final phase of the first Learning in Science Project (F1–4) used an action-research model to develop and research new pedagogies that took into account students’ thinking and a constructivist view of mind (Osborne, Freyberg and Tasker, 1982). This involved local science teachers in the development of viable solutions to the problems identified in the first two phases of the project, and specifically the commonly held alternative conceptions. The action research covered four areas: • •
biology (Bell, 1981a): developing students’ alternative conceptions of living, animal, plant; physics (Osborne and Schollum, 1981): developing students’ alternative conceptions of force, friction and gravity, and electric current;
Pedagogies that take into account students’ thinking
• •
65
chemistry (Osborne, Schollum and Russell, 1982): developing students’ alternative conceptions of burning, reactions, and heating and cooling; classroom activities (Lambert and Tasker, 1982).
The published teacher guides contained a summary of the learning problem; the relevant research findings from Phase 2 on commonly held alternative conceptions of students; background notes on the accepted scientific concepts; a teaching sequence, with teaching and learning activities addressing the alternative conceptions; worksheets; surveys (developed from the interview about instances and events interviews) to assess students’ learning, through pre- and post-assessment; and videos of the teaching sequence being used (Osborne, Freyberg and Tasker, 1982). No evaluation of the learning outcomes was done within the action-research phase of the project. This tended to be done within the associated theses: animal, plant, living (Stead (now Bell), 1980a); photosynthesis (Barker, 1985b, 1986); earth sciences (Happs, 1984) and class activities (Schollum, 1986) and in parallel research projects on mixtures (Cosgrove, 1982a); electric current (Cosgrove and Osborne, 1985a; Cosgrove, Osborne and Forret, 1989; Osborne, 1983). For example, Bell and Barker (1982) documented a research exercise in which a class of twenty-six 13-year-old students were taught by Miles Barker, as was usual, about the biological term ‘consumer’, in conjunction with the term ‘producer’. No teaching was done on the biological term ‘animal’. In contrast, another class (of twenty-four 13-year-old students, had the same teaching about ‘consumer’ and ‘producer’ but were taught the biological concept ‘animal’ first. The results for the second class showed that more students (than in the first class) learnt the scientifically correct concept of ‘consumer’. Teaching which addressed students’ alternative conceptions (in this case of ‘animal’) resulted in better learning of other related scientific concepts. Beverley Bell (1984a), building on earlier work in Stead (now Bell, 1980a) in the third phase of her doctoral research, investigated the effectiveness of three different teaching approaches, each based on Year 9 students reading written texts to learn the accepted scientific concept of ‘animal’. Four different texts were used: •
•
•
The challenge text contained a paragraph which aimed at helping the reader (the student) construct the scientific meaning of animal, that may conflict with the everyday meaning of animal. The dual-meaning text contained sentences which aimed at helping the reader construct a meaning of both the everyday and scientific meaning of ‘animal’. The typical text contained, as did many published texts, sentences about the ecological role of animals, food chains, food webs and about the movement of animals.
66
Pedagogies that take into account students’ thinking
•
The control text was similar to the typical text but the scientific definitions and examples were not elaborated on.
The sample of 227 Year 9 students, were pre-, post- and post-post-tested around the reading of one of the texts, with respect to their concept of animal. Analysis of variance (ANOVA) tests of statistical significance were carried out on the data. More students, after reading one of the four texts, learnt the scientific concept of ‘animal’, particularly with respect to exemplars of the concept. Each of the four text samples gave more correct responses in the post-test than in the post-post-survey but no one text sample returned to the pre-survey results. As a group, the results of the dual-meaning text sample differed significantly from those of the challenge and typical texts. The dual-meaning text sample of students had changed their existing knowledge the least. John Happs (1984) researched the thesis that learning, in the earth sciences, might be better understood if teachers have some appreciation of the learner’s relevant alternative knowledge frameworks prior to instruction. The central aim of this investigation was to monitor the impact of teaching sequences on a particular group of four Year 10 students in the same class, the intervention group, in order to gain some insight into the nature of the interaction between learners and materials, and further insights into the learning process. There were also two control groups (of four Year 8 and four Year 9 students) who had no specific teaching. The 12 students belonged to three classes in the schools. The researcher monitored the learning that occurred for the intervention and control groups in two one-hour visits per term, over 1982 and Term 1, 1983, during which the following data collection was undertaken: • • •
completion of word association survey by pupils; followed by an interview using rock and mineral samples, with the interview transcripts being converted into propositional format and a concept map; completion of repertory grid by pupil within a few days of the interview.
This procedure was adopted for each of the 12 pupils, and repeated twice again for the intervention students following the teaching programme. Following an analysis of the common aspects of students’ alternative frameworks of ‘rock’ and ‘mineral’, a teaching programme over a three-week period (of nine one-hour lessons) was developed with the overall aim of challenging pupils’ alternative knowledge frameworks, providing scientific concepts of rocks and minerals, within a constructivist view on learning. Only the class containing the Year 10 intervention group of students received the teaching programme in Term 1, 1983. The teaching programme was based on the ideas of: • •
finding out about students’ existing knowledge frameworks; consideration of how existing knowledge will interact with given scientific knowledge;
Pedagogies that take into account students’ thinking
• • •
67
pupils being made aware of their alternative frameworks and how these are likely to differ from the views held by scientists; connections being made between scientific concepts and the pupils’ prior knowledge which may or may not be in conflict with the scientific perspectives; pupils being allowed to go beyond the information given. Provision was made for group discussion between pupils so that they can discuss or argue with each other, reconcile differences of opinion or experiment with different points of view. These were termed ‘peer-interaction’ discussions (Happs, 1983).
The teaching programme included such activities as surveys; categorisation activities; practical work investigations; completing crosswords, find the word games, matching items tasks; group discussions; worksheets to complete using short answer sentences; and true and false questions. Further details of the package are given in Happs, 1984, appendix B: 322–444. The evaluations of the learning of the intervention and control groups of students were descriptive, so that the qualitative changes to the students’ conceptual frameworks could be detailed and better understood. The findings suggested that the conceptual development of the control students, without any specific teaching, was minor. For the intervention group, it appeared that the teaching episodes interacted with aspects of students’ existing ideas, resulting in conceptual development. For some students, their existing knowledge was unchanged or altered in unanticipated ways. Some students accepted the new scientific meanings, in part or whole, only for making predictions within a scientific context. Such students tended to shift back to a reliance on their earlier views when interpreting everyday events. As part of his doctoral thesis, Miles Barker (Barker, 1985a; Barker, 1986; Barker and Carr, 1989a, b, c) developed and researched a new teaching package on photosynthesis (Barker, 1985b). The research was in three parts. In the first phase, 28 pupils (aged 8–17 years) were interviewed using interviews-aboutinstances and interviews-about-events techniques to elicit commonly held priorteaching ideas of photosynthesis and plant nutrition. This lead into Phase 2 in which nearly 6000 pencil-and-paper survey responses were obtained from students aged 10–18+ years. In Phase 3, classroom action research was undertaken to evaluate three existing strategies of teaching plant nutrition and photosynthesis: guided discovery, element analysis and trophic conflict strategies, and a new strategy. None of the three existing strategies addressed both the students’ existing knowledge and the scientists’ view, so a new strategy was developed, which explored the material aspects of photosynthesis and which was based on the generative learning model. The new teaching strategy consisted of focus, challenge and application phases and aimed at helping the students construct the scientists’ views of plant nutrition and photosynthesis. The teaching package was shaped according to the eight postulates of the generative learning model, interwoven with the three phases of the generative teaching model. The focus phase activities were to help students become aware of their own ideas about
68
Pedagogies that take into account students’ thinking
plant nutrition; the challenge phase self-teach booklet assisted students to construct the scientists’ view of the material changes in photosynthesis; and the application phase required students to compare their newly constructed scientists’ view with their own earlier views. The new strategy was trialled with a Year 10 science class of 28 students (14-year-olds), over four weeks. In-depth, rich data from multiple sources was collected using observations and audiotaping of teacher/ class interactions; observations and audiotaping of researcher/student and student/ student conversation; out-of-class one-to-one interviews with individual students and the teacher; and analysis of students’ written responses in the investigations, surveys, checkpoints. These data indicated what was happening in the teaching and learning processes in the classroom during the teaching. Pre- and postteaching surveys were undertaken by the students to assess the learning that occurred. The students in classes taught using the guided discovery, biochemical (element analysis) and ecological energetic (trophic conflict) strategies were also assessed for comparison purposes. The results indicated that the students taught by the new, generative learning strategy, but not the usual guided discovery approach, acquired a view of photosynthesis first as a carbohydrate-producing process and also as an energy-storing process and a food-producing process. The action research, related thesis research and accompanying teacher’s guides of the first Learning in Science Project specifically gave teachers information about the children’s science or existing ideas that students brought to the lesson. Two examples of this are the teacher’s guides on burning (Biddulph, 1991) and photosynthesis (Barker, 1985b). However, this pedagogical knowledge has tended not to be included in the support material developed to support the most recent New Zealand national science curriculum (Ministry of Education, 1993b). The new teaching activities arising out of the first Learning in Science Project and associated theses indicated new roles for the teacher which were seen to be wider than merely transmitting scientific knowledge to students, and added the roles of the teacher as: • •
motivator, diagnostician, guide, innovator, experimenter and researcher (Osborne and Freyberg, 1985); creating a supportive atmosphere for learning: – – –
•
personal involvement, valuing students’ ideas, understanding of positive and negative feelings; being a provider of resources, advice, information, for example, the scientific knowledge; helping conceptual change by emphasising study skills, proposing a counterview, tempting further inquiry, reflecting, eliciting ideas, accepting ideas, testing out ideas, linking old and new ideas, promoting new ideas in new situations (Biddulph, 1993);
teaching as researching: finding out what students are thinking; responding: interacting with the students’ thinking; assessing students’ thinking; teacher
Pedagogies that take into account students’ thinking
•
69
as managing learning; teacher as learner; teaching as power on, off, with, for (Bell, 1993b); finding out what the students are thinking; helping students clarify and reflect on their own ideas; challenging students’ ideas; helping students change their ideas; helping students find answers for themselves and ‘getting them to think’; giving feedback; promoting discussion in the classroom; organising social groupings for learning; accepting students’ ideas; creating a supportive, caring atmosphere for learning; motivating and stimulating students to learn; being a co-researcher and learner; providing resources; being a manager and organiser; planning the curriculum (Bell, 1993a).
This indicated the complex and multifaceted nature of teaching, so that the term ‘pedagogy’, a wider, more embracing term, is more useful and appropriate. A main criticism of teaching based on a constructivist view of learning was that the teacher was portrayed as a ‘facilitator’ and not a ‘teacher’. The role of teacher as a provider of information was erroneously seen as teachers not telling and explaining the science to students. But teachers do tell and explain the science to students and in many different ways other than lecturing (Oxenham, 1995). The interaction between teacher and student is not a simple monologue nor characterised by a simple teacher comment and student reply. At the time of the first Learning in Science Project (F1–4), the main debates arising from the newly developed teaching strategies which took into account students’ existing ideas, were as follows: • • • • •
Did students’ alternative conceptions need to be addressed? When and how are the accepted scientific ideas best introduced to learners? Is there a role for conflict in promoting conceptual change and development? The role of discrepant events in promoting conceptual change. The nature and the sequence of the learning task.
This first LISP project established the science education research at Waikato as giving priority to the nexus between theory and praxis, a position that is paralleled by some international literature, as documented in Bell (1993a); Duckworth, Easley, Hawkins and Henriques (1990); Fensham, Gunstone and White (1994); Glynn and Duit (1995); White and Gunstone (1992). The findings of the first LISP project also had implications for the curriculum. As stated previously, the first Learning in Science Project was funded partly in response to the draft 1978 Years 7–10 science syllabus and guide. The research findings had implications for the curriculum, and these were much debated in the following F1–5 Science Curriculum Review (Bell, 1990, 1991b). These implications were as follows (with parallel international publications): •
Science curriculum aims, purposes and desired outcomes (Fensham, 1985; Freyberg and Osborne, 1985). The pedagogical aspects of epistemologies,
70
•
•
•
•
Pedagogies that take into account students’ thinking
views of mind, the nature of knowledge, science curriculum aims and purposes, and roles for the teacher were being problematised. The need for broad learning objectives and the inclusion of thinking skills as learning outcomes was highlighted (Baird, 1984; Bell, 1991a). The format of the curriculum (Bell, 1991a; Driver and Oldham, 1986), for example the inclusion of suggested teaching activities and prescribing of less content to be ‘covered’. There were problems in the way the syllabus and associated written resources were formatted and interpreted by teachers. The science content to be taught and learnt, the science curriculum aims and purposes and the teachers’ professional knowledges were aspects of pedagogy being questioned here. The nature and role of classroom discourse (Bell and Freyberg, 1985; Rodrigues and Bell, 1995), including teachers and students using science language and everyday language in the science classroom. The classroom discourse is an aspect of pedagogical practice that was seen as requiring specific consideration by the teacher, including using a range of classroom activities that enabled differing types of talking and communication in the classroom. The series of narrative science readers (for example, Ministry of Education, 1992a, b, c, d) are examples of such a curriculum resource for teachers of science. The types of activities suggested by teachers’ guides (Bell, 1991a; Driver, 1989), for example the inclusion of pedagogical knowledge of learners’ existing thinking and learning activities that were not the traditional ‘practical work’. The teaching of science in meaningful contexts (Gribble, 1993; Jones, 1982a, 1990; McKinley, McPherson Waiti and Bell, 1992; Rodrigues, 1993b) and the listing of suggested contexts for teaching specific concepts. For example, the draft syllabus (Ministry of Education, 1990a) organised suggested contexts into the contexts of science and myself; the world; leisure and work.
Most of these problems, identified by teachers, students and science educators in 1979–81, were aspects of pedagogy. It is of interest that the pedagogical aspects of assessment and cultural aspects of learning were not mentioned. Another aspect not considered was the nature of science itself. Many of these implications for the curriculum were addressed in the draft F1–5 Science syllabus (Ministry of Education, 1990a) and related teacher guides (for example, Ministry of Education, 1990b) and the 1993 science curriculum (Ministry of Education, 1993b).
The Generative Model of Teaching The Generative Model of Teaching (also called the Generative Teaching Approach) was developed from the work of the action-research phase of first Learning in Science Project. Linked in its theorising to the Generative Learning Model (Osborne and Wittrock, 1985), it included the three teaching phases: focus, challenge, and application, as well as a preliminary phase as indicated in Figure 4.1:
Pedagogies that take into account students’ thinking
71
Teacher activity
Pupil activity
Phase
Ascertains pupil’s views; classifies these; seeks scientific views; identifies historical views; considers evidence which led to abandoning old views.
Completes surveys, or other activities, designed to pin-point existing ideas.
Focus
Establishes a context. Provides motivating experiences.
Becomes familiar with the materials used to explore the concept.
Joins in, asks open-ended personallyoriented questions.
Thinks about what is happening, asks questions related to the concept.
Interprets pupil responses.
Decides and describes what he/she knows about the events, using class and home inputs. Clarifies own view on the concept.
Challenge
Interprets and elucidates pupils’ views.
Presents own view to (a) group (b) class, through discussion and display.
Facilitates exchange of views. Ensures all views are considered. Keeps discussion open.
Considers the view of (a) another pupil (b) all other pupils in class, seeking merits and defects.
Suggests demonstrative procedures, if necessary.
Tests the validity of views by seeking evidence.
Presents the evidence for the scientists’ view.
Compares the scientists’ view with class’s view.
Accepts the tentative nature of pupils’ reaction to the new view. Application Contrives problems which are most simply and elegantly solved using the accepted scientific view. Assists pupils to clarify the new view, asking that it be used in describing all solutions.
Solves practical problems using the concept as a basis.
Ensures students can verbally describe solution to problems.
Presents solutions to others in class.
Teacher joins in, stimulates and contributes to discussion on solutions.
Discusses and debates the merits of solutions, critically evaluates these solutions.
Helps in solving advanced problems; Suggests further problems arising suggests places where help can be sought. from the solutions presented.
Figure 4.1 The Generative Model of Teaching (Cosgrove and Osborne, 1985a: 109)
72
Pedagogies that take into account students’ thinking
A
Model A: There is no electric current in the wire attached to the base of the battery.
B
Model B: The electric current is flowing in a direction toward the bulb in both wires.
C
Model C: The direction of the electric current is one way. The current will be less in the return wire.
D
Model D: The direction of the electric current is one way. The current will be the same in both wires.
Figure 4.2 Children’s models of electric current (Osborne and Freyberg, 1985: 23–4)
This teaching model was used to teach the concept of electric current (from technological contexts) (Cosgrove and Osborne, 1985a, b; Osborne, 1983). Previous research had described the four views of electric current (see Figure 4.2), commonly held by school students (Osborne, 1981; Osborne and Gilbert, 1980b). Views A, B, and C are alternative conceptions and view D the scientists’ view. These results are similar to those reported elsewhere (Shipstone, 1984).
Pedagogies that take into account students’ thinking
73
A teaching package, designed to promote the conceptual development of the students’ alternative views (A, B, C) towards the scientists’ view (D), was developed, trialled and researched by Mark Cosgrove for his doctoral research and documented in Cosgrove, 1989; Cosgrove and Osborne, 1985a, b. A key research question was: does this approach shed light on what learners can do (as distinct from whether or not they get the ‘right’ answer)? The teaching model was linked to the then national syllabus (Department of Education, 1978) requirements on electric current in Years 7–11. The ‘Generative-Learning’ Teaching model (the term Cosgrove used), in this instance, can be described as a conceptual change model, which featured problem solving, addressed students’ existing ideas of electric current, and comprised the focus, challenge and application phases. Of note is the use of a critical experiment in the challenge phase. In this ‘critical experiment’, the students used ammeters in different parts of the circuit to confirm (if they had the currently accepted scientists’ view D) or challenge their views of electric current (if they held an alternative conception (views A, B, C). Qualitative data-collection techniques – observations, clinical interview (the interview about-events), general interviews with learners and teachers, and surveys – were used to research and evaluate the teaching model with respect to the learning process at the different phases, the learning outcomes, and the benefits and disadvantages of using the model as perceived by the teachers and students. Pre-, post-, and post-post evaluations were done of the students’ views of electric current. The teaching approach and package was trialled on six occasions with different teachers and classes, with students in Years 7 and 10; in girls’, boys’, and co-ed schools. The learning during the teaching was also researched in depth over a five-day period. The study confirmed that many learners have prior ideas about electricity which are undifferentiated with respect to current and voltage, and have based their thinking on sequential reasoning and on existing ideas about consumption (‘getting used up’). The results indicated that some learners adopted new views initially but regressed to their earlier views. Others adopted a new view, and for some this resembled the scientists’ notion that electric current is conserved. Significant changes occurred where learners were able to invent analogies and explore the implications of these analogies in a relaxed environment where there was plenty of time. Learners were not able to change from the sequential reasoning, nor from their belief that something is consumed; they coped with the latter idea by proposing a two-component system in which one component, the current, is conserved but the other, a ‘fuel’, is not. Changing to a two-component model, called a transitional framework, is thought to be as big a differentiation as can be achieved. It is unlikely that learners would develop a view of electric current consistent with scientific ideas at first contact. Girls appeared not to be disadvantaged by this approach. A handbook for teachers and a student’s workbook were produced (Cosgrove, Osborne and Forret, 1989).
74
Pedagogies that take into account students’ thinking
The Generative Teaching Model, with its three phases of focus, challenge and application, has been researched with secondary teachers and learners of science, for example in Barker, 1986; Jones, 1988; Mohamed, 1995, and used as a basis in the teaching guides produced at Waikato, for example Barker, 1985b; Biddulph, 1991; Cosgrove, 1982b; Cosgrove and Meuggenberg, 1986; Jones, 1990.
The Interactive Teaching Approach The Interactive Teaching Approach (Biddulph and Osborne, 1984), which addressed the pedagogical role of students’ questions in the classroom, was developed in the Learning in Science Project (Primary) (Biddulph and Osborne, 1985; Osborne and Biddulph, 1985b). It was a teaching approach based on a constructivist view of learning that took into account students’ thinking, including their questions (Biddulph, 1989, 1990b; Biddulph, Osborne and Freyberg, 1983b; Harlen and Osborne, 1983; Osborne and Biddulph, 1985a). However, the questions were not seen as teachers asking questions to which they already knew the answer, but as student-initiated questions to facilitate their learning. The argument or thesis was that primary school children’s own questions about natural and technological phenomena could have considerable significance for their learning in science. Fred Biddulph (1989), in his doctoral thesis, argued that a pedagogy to promote learning of science using the children’s own questions could be justified on curriculum, pedagogical, philosophical and psychological grounds. The pedagogical grounds included the teacher obtaining insights into what the students were thinking, and using the questions to start where the children are at. The key parts are: preparation; exploratory activities; children’s questions; children’s investigations; and reflection (Biddulph, 1990b), as illustrated in Figure 4.3. The use of the term ‘interactive’ to describe a teaching approach is of importance here. The first LISP project had identified classroom communication as important in promoting conceptual development. The teaching approach developed in the second project promoted classroom discussion and dialogue that differed from the normal pattern of ‘teacher question–student answer–teacher response’. The classroom discussion promoted by the interactive teaching approach was interactive: students and teachers were interacting in talking with each other. In other words, students’ and teachers’ minds were interacting. This was to become one of several themes in the continuing LISP research programme at Waikato. Examples of an Interactive Teaching Approach were documented in a collation of teachers’ guides (Biddulph and Osborne, 1984). These guides contained scientists’ ideas for teachers on the topic to be taught and learnt; possible students’ questions and existing ideas and explanations; teaching activities, for example exploratory activities, specific investigations that students can carry out;
Pedagogies that take into account students’ thinking PREPARATION The teacher and class select the topic and find background information.
BEFORE VIEWS The class or individuals say what they know about the topic.
EXPLORATORY ACTIVITIES Involve the children more fully in the topic. COMPARISON CHILDREN’S QUESTIONS A time when the class is invited to ask questions about the topic.
INVESTIGATIONS Teacher and children select questions to explore, say 2 or 3 per day, over a 3 or 4 day period.
MORE QUESTIONS
AFTER VIEWS Individual or group statements are compiled and compared with earlier statements.
REFLECTION A time to establish what has been verified and what still needs to be sorted out.
Figure 4.3 The Interactive Teaching Approach from (Faire and Cosgrove, 1988: 16)
75
76
Pedagogies that take into account students’ thinking
learning activities for metacognition; and resource material. In other words, the guides contained a wide range of information relating to pedagogical practice; primary science curriculum aims, purposes and desired outcomes; science content to be learnt; science classroom discourse; science teacher knowledges about pedagogical content knowledge, learners and the nature of knowledge; teaching strategies and skills; and science learning activities and resources. Of interest was the emphasis given to the new roles for the teacher in teaching science. These roles were as a stimulator of curiosity, a challenger of ideas, a resource person and senior co-investigator (Biddulph, 1990a) and were in addition to the role of the teacher as communicator of the scientists’ views, which was downplayed but not absent. The role of the resource person was suggested to communicate to teachers that they could supply the scientists’ ideas to the classroom debates, or use a book to input the scientists’ ideas or invite an expert into the classroom to do so. This was to remove the perceived burden on primary teachers of science (who tend not to be specialists in teaching science) that they had to know all the answers before they could teach science. The evaluation of the teaching approach was through qualitative self-report data from the teachers and pupils who had used the interactive approach (Biddulph, 1985, 1990b). It was widely reported by the teachers and students that they were positive about using the approach, ‘in terms of its benefits to children’ (Osborne and Biddulph, 1985b: 55). The teachers reported (Biddulph, 1985, 1990b): • • • • •
• •
increased interest by students; increased happiness, busyness, cooperation and engagement of students; increased student involvement in the investigations; increased student curiosity, willingness to communicate their ideas, openmindness; the teachers’ increased attention to help students develop their questioning, investigatory and communication skills, and ‘that these improved remarkably’ (Osborne and Biddulph, 1985b: 57); increased development of (process) skills, for example, observation; increased development of intellectual skills.
The students reported that they preferred learning science from their own questions as it was more interesting, more challenging and because they learnt more (Osborne and Biddulph, 1985a). The self-report data from the teachers (Biddulph, 1985, 1990b) indicated that using the interactive teaching approach had: • •
increased their confidence in teaching science; provided them with strategies to listen to and work closely with their children so that they became far more aware of what it was the children were bringing to the lesson, and gaining from a particular lesson;
Pedagogies that take into account students’ thinking
• • •
77
made the children’s real abilities evident; helped develop a sense of ‘community’ in the classroom; enabled them to cater for children of varying ability.
These gains helped the teachers feel better about themselves as teachers, and they reported that they used the approach in other curriculum areas (Biddulph, 1985). The self-report data indicated that the teachers also encountered difficulties, including: • • • • • • • •
dealing with the superficiality of the children’s initial questions; thinking of investigations for some questions; the diversity of questions; the change in the teacher’s role; the intensive nature of the teaching; finding suitable resources; assessment of learning; the children’s lack of investigatory and meta-cognitive skills (Biddulph, 1985).
While the evaluation did not provide evidence of students’ increased learning outcomes per se, the self-report data did provide evidence of ‘better learning conditions’ (Bell and Pearson, 1992a). These ‘better learning conditions’ were pedagogical signs to the teachers in the classroom that better learning outcomes were more likely to be obtained. This view parallels the prevailing one at the time that an important role of the teacher is to structure the learning environment. Accounts of teachers who have used the Interactive Teaching Approach can be found in Bell, 1993a. A further primary science teachers’ guide (Faire and Cosgrove, 1988) and a secondary teachers’ guide for teaching about earthquakes (Hume, 1992), both using the Interactive Teaching Approach, were also produced. Another type of evaluation within the LISP (Primary) was how the teachers used the teacher guide material and the interactive teaching approach (Biddulph, 1985). This evaluation was in line with the focus of the project on ways to help teachers develop their teaching to take into account the findings of the research on how children think, and the ideas and explanations they bring to the science lesson (Symington and Osborne, 1984). This theme was to continue to be developed in later Learning in Science Projects. The evaluation indicated that teachers who did not already operate within the constructivist theoretical framework could not take the written materials and implement them as intended. These teachers needed considerable in-service or pre-service support to develop their views of learning, before they could successfully use the approach. The use of the Interactive Teaching Approach by six beginning primary teachers of science was also investigated (Fernandez, 1991; Fernandez and Ritchie, 1992). Interviews revealed that the teachers initially faced several obstacles to the
78
Pedagogies that take into account students’ thinking
implementation of the Interactive Teaching Approach, which they had first encountered in their pre-service teacher education. These included lack of collegial support, lack of feedback on their teaching, difficulty assessing the learning of their pupils, and differences between the culture of learning of the Interactive Teaching approach and that of the pupils. However, by the end of the year, the teachers had reconstructed the teaching approach in ways that reduced these difficulties. An ongoing concern was that of keeping a focus on the science to be learnt, whilst using the Interactive Teaching Approach. For some teachers using the interactive approach, the ‘science’ became lost in the sea of language activities in the context of a topic such as spiders, rocks, metals. This developed into a debate on the role of the teachers’ pedagogical content knowledge.
The content of science Another key finding of the LISP research programme has been the importance of the content in a debate on pedagogy and learning. The research on the role of students’ existing ideas has indicated not only how the science content in part shapes the teaching and learning processes as in traditional curriculum planning, but also how the pedagogy and learning processes shape the content learnt. The teachers’, students’ and scientists’ existing views of a concept (for example, of energy or animal or boiling) interact in the communication during teaching and learning to influence the learning outcomes strongly. The content to be learnt is not unproblematic, nor is it a generic, neutral aspect of pedagogy. The scientists’ concept of ‘energy’, for example, is not transmitted from sender to receiver unchanged. As the sender and the receiver both construct understandings during the communication process and negotiate shared understandings there is much room for miscommunication and unexpected and unintended learning outcomes. For example, a teacher may say that water is made of hydrogen and oxygen and understand it to mean it to be a molecule comprising hydrogen gas and oxygen bonded chemically together. But the students who say that the bubbles in boiling water are bubbles of hydrogen and bubbles of oxygen gas may or may not have the notion of their being chemically bonded. The content (water is made of hydrogen and oxygen) is not unproblematic. The role of content in the teaching and learning of science was addressed in the third LISP project. The Learning in Science Project (Energy) was a three-year project spread over the years 1985–88 (Kirkwood and Carr, 1988a, 1989) to investigate the teaching of the concept of ‘energy’ to 5 to 18-year-olds. (Refer back to Chapter 1 for further details of the project.) This research is most notable for the determining influence of content on these pedagogical aspects. This strength of the LISP (Energy) project is discussed in the following sections. The difficulties experienced by primary and secondary students and teachers in learning and teaching the concept of ‘energy’ were explored in the existing literature (Welch, 1985) and with local teachers of science (Kirkwood et al., 1985).
Pedagogies that take into account students’ thinking
79
There was a general consensus that science educators and students felt the concept of ‘energy’ to be abstract and difficult to teach and learn. The scientists’ view of ‘energy’ was felt to be vaguely defined and prescribed in the curricula and so the research team developed guiding principles which were ‘intended to assist teachers at all levels and in all disciplines in their consideration of energy topics. They were not intended as a definition of energy to be taught to students’ (Kirkwood et al., 1985: 8). These guidelines were in essence an aspect of pedagogical content knowledge and an attempt to structure the content for ease of use by teachers to address the difficulties faced by students in learning about ‘energy’ (Carr, Kirkwood, Newman and Birdwhistell, 1987a, b). They were: 1 2 3 4
Energy is seen clearly focused only when a defined change to a system is being considered. In defining this change, the boundaries or limits to the system need to be described. Only then can descriptions of changes be fruitful. When considering defined changes to defined systems, the initial consideration should be of observable and measurable changes. The principle of conservation of energy requires that when energy changes in form, the total amount of energy remains constant. (Kirkwood et al., 1985: 8–9)
These were later developed and revised to: 1
2
It is necessary to analyse systems undergoing change to clarify the concept of energy. 1.1 the system may be open or closed. For such an analysis of a system undergoing change, it is necessary to describe the initial and final conditions by specifying for the components their: (i) (ii) (iii) (iv) (v) (vi)
3
4 5
relative positions; chemical composition; temperature; speed; state (solid, liquid or gas); mass.
To explain the changes described by comparing the initial and final conditions of the system, and to more fully describe the change, it is necessary to introduce the concept of ‘energy’ as a non-material factor in the change. Energy exists in a variety of forms, e.g. light, sound. When energy changes in form, the total amount of energy remains constant. (Kirkwood, 1988: 7)
80
Pedagogies that take into account students’ thinking
From the start of the project, the influence of the content on the pedagogy was acknowledged. The way the content was structured and conceptualised was seen as determining the pedagogy. The existing ideas that students and teachers bring to the science lesson were also investigated, building on the findings on ‘energy’ in the in-depth phase of the first Learning in Science Project (Stead, 1980b), and confirmed the international literature of children’s views of energy (Welch, 1985). It was reported (Kirkwood and Carr, 1988a) that the alternative conceptions of ‘energy’ held by students of all ages were that ‘energy’ is: • • • • • • • • •
a general kind of fuel that does much work for us; associated with living things; associated with moving things, e.g. fire, car; able to take on different forms as it travels through wires or chains on bicycles; a source of force or activity stored in objects, e.g. water has energy in it so it can turn a water wheel; a storehouse used to make things work such as a battery; able to be obtained from food, the body, sun and soil: it is an ingredient stored in them; a fluid-like material that flows from one body to another, as an electric current or stream; given off like a waste product, for example, chemicals give off heat. (Kirkwood and Carr, 1988a)
The most notable feature of these alternative conceptions is that energy is viewed as a substance by many students of all ages. It was also reported that students have difficulty understanding the concept of energy conservation but not those of energy transfer and energy flow. Students also demonstrate considerable confusion between the word ‘energy’ and other terms such as force, power and work (Kirkwood and Carr, 1988a). It was also reported that science teachers differ in their views of ‘energy’. Biologists tend to view ‘energy’ as what is required by living things for the maintenance of life and which flows through the ecosystem. Chemistry specialists tend to hold a view associated with the rearrangement of bonds in matter and the physics specialists’ view is that energy is the ability to do work (Kirkwood, 1988). Hence, both students and teachers bring to the lesson differing views of ‘energy’, some of which are alternative ones to the accepted scientific ones. This may cause considerable confusion for students in their junior secondary schooling when they may have teachers of different specialities. The influence that these existing ideas have on the learning outcomes was also investigated. As indicated above, students aged 5–17 held views of energy that are alternative to the views of scientists. In other words, school teaching on
Pedagogies that take into account students’ thinking
81
‘energy’ in primary and secondary science lessons had not resulted in the students developing their concepts of energy into the accepted scientists’ views. The LISP (Energy) project detailed some of the difficulties in teaching and learning ‘energy’ at the junior secondary level (Carr and Kirkwood, 1988; Kirkwood, 1988). These were: • • • • • • • • • • •
confusion in the classroom as to what constitutes a form of energy and about different forms; conceptual and semantic difficulties with potential energy; confusion between energy forms and energy resources; energy ‘seen’ in static situations/objects with what happens next implicit; the use of energy as a signal; difficulties between the everyday use of the phrase ‘conservation of energy’ and the scientists’ conservation of energy; confusion about energy conservation and energy degradation; unhelpful and confusing use of extra words, e.g. direct/indirect; renewable/non-renewable; inconsistent view of energy portrayed in teacher resource material; implicit and inconsistent use of systems by teachers; unawareness of students’ ideas and few links made to their ideas about energy. (Kirkwood and Carr, 1988a: 18)
It was felt that these difficulties were characterised by a lack of teaching that takes into account students’ thinking and a lack of classroom discussion, dialogue and interaction on the meanings and understandings being constructed in the classroom by teachers and students. A draft teaching package (in Kirkwood, 1988), promoting teaching that takes into account students’ thinking on ‘energy’, was developed and trialled by the researchers (Kirkwood, Carr and Newman, 1988) and evaluated in an actionresearch mode (Kirkwood and Carr, 1988a). A teachers’ guide based on this package (Kirkwood, 1989) was then produced and distributed to every secondary school in the country. The main points addressed in the alternative teaching material included: • • •
•
the open acknowledgement of the invented character of the concept of energy; the concept of potential energy is not helpful at this level; the concept of energy is clearer for students when it is taught in the context of defined systems undergoing change involving familiar objects as far as possible; in defining the system being studied, teachers should discuss with students whether people are part of the system and negotiate boundaries to these systems with students;
82
Pedagogies that take into account students’ thinking
• • •
a distinction is made between the transformation and transfer of energy; taking into account the everyday meaning of ‘conservation of energy’; avoiding the use of the term indirect/direct; primary secondary; renewable/non-renewable. (Kirkwood and Carr, 1988a)
To enable these teaching aspects to be carried out in the classroom, different teaching and learning activities were used, including such techniques as brainstorming, small-group discussions, posters of students’ ideas, opened class debates. These alternative teaching approaches were demonstrated to be effective for students’ learning about the concept ‘energy’ as illustrated by the students’ and teachers’ self-report data and pre-/post-survey results (Kirkwood, 1988). The pre-/post-survey results for the three classes (n=28; n=34; n=25) showed more development of the students’ ‘energy’ concepts than one might expect given the results of the cross-age studies of children’s concepts of energy (e.g. Stead, 1980b). The three teachers and the researcher involved in the research (self-) reported considerable gains in their professional development as a result of being involved in the research project (Kirkwood and Carr, 1988b). All participants felt that they had started a process of change; had developed an awareness of their own and students’ views of ‘energy’; had developed a greater awareness of their own and others’ teaching styles; were now able to critique the then currently used assessment methods; had reflected on their teaching and science; and had developed alternative teaching strategies, recorded in curriculum materials. It is not known whether these teachers still use these alternative teaching strategies with the 1993 science curriculum (Ministry of Education, 1993b), nor whether teachers outside the research project adopted the strategies. Therefore, a teaching approach that took into account students’ prior ideas and their understandings constructed during the lesson, and that kept a focus on the science to be learnt, was shown to be more effective than traditional teaching approaches. The degree of professional development required by teachers to teach the new approach was considerable. In summary, the content to be taught and learnt determined the pedagogical aspects of teaching and learning activities, classroom discourse, progression and the relevant pedagogical content knowledge.
The contexts of teaching science The importance of ‘context’ was also raised by the development of a constructivist view of learning and mind at Waikato as well as internationally (Cognition and Technology Group at Vanderbilt, 1990). Learning is not context-free; it is embedded in social and cultural contexts (Bell, 1993a). Hence, the pedagogical importance of considering the context is not just for motivational purposes but it is also an important part of the learning process. In constructing an under-
Pedagogies that take into account students’ thinking
83
standing, learners must make links between their existing knowledge and the new knowledge being taught in the science classroom (Bell, 1984a; Osborne and Wittrock, 1985). Also of pedagogical importance was the role for the teacher of helping students to make these links ( Jones, 1982b; Osborne and Freyberg, 1985). The LISP research programme has researched the use of contexts that have meaning for the students, as an effective pedagogy for science education. These contexts are those that students find interesting and relevant, could relate to, ask questions about, develop and use to explore their own ideas and make sense of their world. Such contexts enable students to link their existing ideas, experiences, values and cultural experiences to the science to be learnt. Other contexts may be used to help students use their newly learnt science with confidence. The findings indicated that teaching science in contexts that enable students to make these links, between the familiar and the unfamiliar, results in higher science learning outcomes. These findings are now detailed. Alister Jones ( Jones, 1988; Jones and Kirk, 1989; 1990a, b) researched the use of technological contexts for the teaching of capacitance and the Doppler Effect with Year 13 physics students’ learning, based on earlier work ( Jones, 1982a, b; Jones and Osborne, 1985). The two conceptual areas were chosen as they were included in the national senior physics examination prescription. Initially, 40 students were interviewed and 500 students surveyed to ascertain their existing ideas, including interests, with respect to technological applications. The results indicated that students were interested in applications within their own environment, directly involving people and aspects which corresponded to their intended careers and anticipated needs. There were gender differences in interest. Students were generally not interested in ‘school’ physics or domestic applications. Following this initial research, the possible ways of introducing technological applications into senior physics were examined in a small study. A new teaching strategy, based on the generative learning model, was developed for the two Year 13 (age 16–17 years) topics of electrical capacitance and the Doppler effect, based on the generative model of teaching (Cosgrove and Osborne, 1985a, b). The teaching sequence consisted of five stages: 1
2
Focusing: to focus the students’ attention on the learning situation in terms of a technological application, a social issue, or an instance that relates to the students’ existing knowledge. A context for learning is established and students are encouraged to ask questions, put forward tentative statements, provide further information and discuss their ideas with others in the class. Exploration: to carry out investigations, to try to answer questions they may have and further explore their tentative statements. The further investigations might be experimental, demonstrations, reading activities, group and class discussion, comprehension exercises, and technological and theoretical problems. The students can explore in more detail how devices work and the physical principles behind them. Historical, social and everyday aspects may also be examined.
84
Pedagogies that take into account students’ thinking
3
Reporting: the students report back to the class the results of their investigations, discussing ideas and making summaries of the physical concepts involved. Formalisation: the teacher and students use the information gained so far to develop physics concepts and write them in a formal physics manner. Application: students may carry out further investigations, do traditional school physics and technological problems.
4 5
The teaching sequence for each of capacitance and the Doppler effect were designed to fit within the normal curriculum time frame of 10 one-hour periods. The teaching packages provided included a teacher’s guide, student reading material, models of technological devices, photographic slides, student experimental kits, student experiments and video footage of technological applications. The contexts used were technological applications of the science ideas being taught and learnt: the camera flash, the microphone, the Apnoea mattress, altitude measurement, the thickness measurer, a humidity sensor, and the early detection of volcanic eruptions. This strategy was initially evaluated with two classes (with a total of 33 students) for the capacitance and three classes (with a total of 52 students) for the Doppler effect, and in the classes of other teachers in other schools. The data collection techniques included classroom observations during the lessons; interviews with teachers before, during and at the end of the teaching sequence; interviews with students following the lessons (usually the next day) and at the end of the teaching sequence. The students were also interviewed 4–6 weeks later to investigate their understanding and recall of the unit. The students were also interviewed using a semantic differential-type scale as a probe to ascertain their perceptions of the unit. The scale consisted of 16 bipolar adjectives developed by Roger Osborne (1976) and was used as a pre- and post-teaching data collection technique. The students were also interviewed individually with both completed scales before them to elicit the reasons for their responses. The student interviews were carried out with the trial class and with a class that had been taught in a more traditional manner. Further trials were carried out in other schools. The evaluation consisted of analysing the pre- and post-teaching data of the students’ perceptions, using a semantic differential-type scale, with a two-tailed t-test to calculate the change in mean response to each item. The self-report data in the interviews was also used. No pre- and post-testing of learning outcomes was done. The findings indicated that, compared to previous teaching programmes, the students were generally very positive about the approaches. The reasons the students gave for being more positive were: the introduction of technological applications that they were interested in and could relate to, the experiments, individual projects, the class discussions and being able to explore ideas for themselves. They also reported that they were more confident to attempt traditional physics problems. Students also found the lessons interesting, easier to
Pedagogies that take into account students’ thinking
85
understand, enjoyable, non-mathematical, technological rather than theoretical, more straightforward and relevant to everyday life. They found the physics concepts easier to remember and were able to relate to both school physics and technological problems. Students who self-reported that they had been bored with previous physics lessons reported that they became more interested and increased their contributions to classroom discussions. However, some students who were more concerned with formulae and answering theoretical problems, as requested in national examinations, did not respond favourably to the teaching sequences. A teachers’ guide was also produced ( Jones, 1990). Susan Rodrigues (1993b) developed and researched a programme of work based on contexts that were meaningful for Year 12 chemistry students in a unit of work on oxidation and reduction (Rodrigues, 1993a). The contexts included the following: •
•
•
•
Thiobaccillus bacteria and the New Zealand farmer. These bacteria found in New Zealand soils convert sulphur to sulphates at ambient temperatures and pressures. A class discussion involving a comparison between this scenario and the ‘contact’ process was proposed. Hair today . . . gone tomorrow. This unit investigated the chemistry of hair perming and colouring, starting with a newspaper article about the pop star Madonna. I can Al Right: a role play in which five groups of people (environmentalists, local people, scientists, company personnel and the local council) consider the development of an aluminium smelter in their town. Driving teenagers to drink. This unit asked whether there should be a zero breath alcohol level for teenagers. The students were then asked to design a breathalyser.
The units were used over a period of 10 lessons in a class (of one teacher) of 20 Year 12 (aged about 16 years) female students in an urban single-sex high school. While one focus of the research was on the learning of science by girls, the research also produced findings on the learning outcomes from a context based senior chemistry teaching unit. In the first phase, 11 units were developed and trialled to identify appropriate contexts. The trial involved 69 senior chemistry female students, from four classes in three schools, responding through pre- and post-questionnaires, and evaluation slips. In the second phase of the research, six of the teaching units were used in one class of 20 senior chemistry female students. The teaching and learning was monitored by audiotaping of the lessons, mid-intervention letter, pre- and postquestionnaires, and concept maps. Four students were interviewed, with audiotaping, after each lesson. In Phase 3, the researcher used three of the units with a class of six female students, with data being collected using audiotaped sessions, pre- and postquestionnaires, journal entries and a post-interview.
86
Pedagogies that take into account students’ thinking
The findings indicated that the students were able to identify the chemistry in each lesson and perceive a sequence linking the units, and they added to their existing knowledge of oxidation and reduction. The students also indicated that the units were relevant and interesting, and provided them with an opportunity to discuss ideas. The students also commented on the units having increased their motivation to learn and making them think. Specifically, the findings indicated that learning redox concepts in meaningful contexts: • • •
required students to take charge of their progress; resulted in students proffering support and providing encouragement; resulted in a loss of teacher-orchestrated turn taking (in classroom dialogue), and generated conversation talk patterns in the classroom; • resulted in student talk being a mixture of everyday language and chemical jargon; • fostered student construction of questions, proposals of new questions, setting up hypotheses in the form of questions; • resulted in a sharing of expertise between the teacher and students; • encouraged students to challenge each other and the teacher; • caused students to volunteer evidence in the form of existing experiences to validate or dispute ideas; • fostered links between existing student experiences and new experiences; • made students reflective when working in context; • resulted in students interrelating viewpoints; • still enabled students to see the chemistry present within the context and a (science conceptual) sequence between the contexts; • enabled students to transfer and use information from one context to the next; • brought to light some misconceptions held by students which the teacher was able to address; • enabled students to modify and add (in various degrees) to their knowledge of redox concepts. These aspects resulted in more understanding and recall of redox concepts by the vast majority of students in Phases 2 and 3. (Rodrigues, 1993b: 266–7) Wayne Gribble (1993) used the contexts of Maori myths and legends, and specifically the two legends of ‘The Warrior Mountain’ by Katerina Mataira and the Legend of Whakaruamoko, during the teaching of a unit on volcanoes and earthquakes to two Year 9 classes. This research is discussed in detail in Chapter 8. Abdul Muhsin Mohamed (1995) explored the processes and outcomes of using technological contexts that are relevant to students for the teaching of senior secondary school physics in the Maldives (a non-western country) and New Zealand (a western country). The research was conducted in four iterative phases. The first phase explored technological contexts which two groups of students (in
Pedagogies that take into account students’ thinking
87
the Maldives and New Zealand) considered relevant. Students in the sample of boys and girls from Grades 10 and 11 (senior) physics classes in the Maldives (n=32) and the corresponding Years 11 and 12 (senior) physics classes in New Zealand (n=32) were individually interviewed to clarify their interest and ideas about applications used in an area of technology common to the two countries – telecommunications. Despite the sociopolitical and economic differences between Malé and Hamilton, the provision of telecommunications was similar at the time of the study, e.g. International Direct Dialling telephone services, card phones, coin phones, satellite TV in addition to local TV, local radio broadcasting on FM and AM, and facsimile. The similarity in student interest in the two groups of students indicated that similar contextual teaching strategies may be successfully used in senior physics classes in the two countries. A classroom intervention was developed to teach a common curriculum area in both countries – electromagnetism – with specific learning outcomes in the curriculum targeted. The context chosen was the transfer of sound in telecommunications. The framework of the teaching unit was that of the Generative Teaching Approach (Cosgrove and Osborne, 1985a), having a focus phase: to engage students with the relevant contexts through classroom discussions and explorations; challenge phase: to engage learners with specific aspects of the contexts and related physics concepts; and an application phase: to apply the newly learnt physics concepts in other contexts. The teaching unit was used in the Maldives and New Zealand in 1993–94 and in its final version for 14 onehour sessions. Phases 2, 3, and 4 were the researching of a classroom teaching intervention, with Phase 3 focusing on the teachers’ views of the intervention. One of the research questions, of interest here, is ‘What are the affective and cognitive outcomes of using the contextual learning and teaching strategies in mixed ability classrooms?’ The pre- and post-testing of the learning outcomes (both cognitive and affective) was done using concept mapping, and conceptual tests, with informal conversational interviews; observations of classroom sessions; questionnaire with informal conversational interviews; and the unit test conducted by the school was used as a post-post-test. The findings of Phases 2 and 3 of the study indicated that the teaching intervention enhanced the affective and cognitive outcomes for the students who previously did not perform well in physics classes. The high achievers continued to perform well and they too were motivated and interested during the contextual teaching. There were no significant gender differences. (p. 163) Rachel Wood (1996) suggested the contexts of Supertoms, AIDS, making babies and fingerprinting and Darwin today to teach genetics to students between Years 8 and 13. This thesis argues for the development of a learning and teaching
88
Pedagogies that take into account students’ thinking
approach which builds on learners’ prior views, using relevant contexts to promote conceptual evolution. The approach had three essential features: the incorporation of student ideas; the use of relevant contexts for learning; and a structure which facilitates conceptual evolution. The research issues were: 1 2 3
What do young students think about basic genetics? How does this thinking change as they study genetics at school? What are the implications of that thinking for their understanding of genetics concepts?
The methodology was in three parts: 1
2
3
The elicitation of students’ prior views about genetics. Three samples were involved: 30 students in Years 6–8; one class of 28 Year 10 students; 17 older students – Years 11–13. The teaching approach was developed, with the three features (the incorporation of student ideas; the use of relevant contexts for learning; and a structure which facilitates conceptual evolution) designed to recognise the mismatch between the school and student perspective. The validation of the teaching and learning approach by a group of five experienced teachers, doing a ‘reality check’. The teaching approach was critiqued but not trialled. The approach was designed to promote effective learning and teaching by: • • •
•
concentrating on a few significant concepts within each context to provide time to explore both the school and student perspective; recognising the factors of: individuality of learning; learning as influenced by social values; science as shared social knowledge; including the three essential features of the incorporation of student ideas; the use of relevant contexts for learning; and opportunities for conceptual evolution; promoting the development of personal values and knowledge, appreciation of social debate on science issues, access to the science culture.
David Porteous (1997) investigated meaningful learning ‘contexts’ in senior physics education. Questionnaire surveys were designed to identify the reasons why students were choosing to study physics, and what learning contexts interested them. Another questionnaire survey for the physics teachers focused on their reasons for choosing particular learning contexts, which is a feature of the then new physics curriculum. The teachers’ views about teaching physics by contextual methods were investigated by way of interviews. The 590 participants in the research comprised 18 physics teachers and 572 physics students (170 female, 402 male) in nine schools, in Years 11–13.
Pedagogies that take into account students’ thinking
89
Students’ and teachers’ choices of contexts did not always coincide. Both selected learning contexts according to their interests and relevance to their lives or the curriculum. The teachers were positive about the implementation of the new curriculum using partial or total learning contexts. Their concerns involved available resources, the time issue, assessment, the gender issue, the changing composition of the physics class, the need for professional development, and whether contextual teaching was appropriate for senior examination classes. In summary, teaching science, in meaningful contexts for students, has been an emphasis in the 1980s and 1990s New Zealand national curricular developments. Research undertaken at Waikato has indicated the benefits to both the learning conditions and outcomes (Bell and Pearson, 1992a). However, concerns have been raised with respect to the size and scope of a suitable context; where contexts might be introduced into a lesson sequence; and clarifying and highlighting the science to be learnt (Hipkins and Arcus, 1997). In summary, the Waikato research into pedagogies for learning science can be characterised as researching teaching that takes into account students’ thinking, and which is theoretically closely linked with learning theories. Over the twenty years, the research has investigated, developed and refined pedagogies that address students’ thinking, including both eliciting student thinking and how we as teachers respond to help the students learn science. Further research on pedagogies for teaching and learning science are reported in the next chapter.
90 More pedagogies Chapter 5
More pedagogies
In this chapter, a review of the Waikato research on pedagogies for science education is continued, namely: • • • • •
Practical work; Inclusive pedagogies; Metacognition; Classroom discussion; Teacher education and development for new pedagogies.
Practical work Practical work, as a pedagogy, has also been researched at Waikato. It started with the second phase of the first Learning in Science Project which explored in depth the problems perceived with the ‘practical’ work in science lessons (Tasker, 1980; Tasker and Freyberg, 1985; Tasker and Osborne, 1983, 1985; Schollum, 1986). In this research, there were two kinds of portrayals of children’s classroom experiences. One was the examination of a few lessons in considerable depth (Osborne and Tasker, 1980; Osborne, Tasker and Stead, 1979; Tasker and Osborne, 1982a, b, c). The second kind of portrayal of children’s classroom experiences was a much more detailed analysis of some 40 lessons (Tasker, 1981; Tasker and Freyberg, 1985). The main finding was the differing perceptions that teachers and students have of the same classroom experiences. The differences were in terms of the scientific context of the activity, the scientific purpose of the activity, the scientific design of the investigatory activity, doing the activity, getting the results, thinking about what was done and what happened, the impact of the experience on children’s views and the relationship to predetermined outcomes (Tasker and Freyberg, 1985). This aspect of the research addressed many key aspects of science pedagogy, in which the need for and the role of practical work was an unchallenged given. Tasker’s work highlighted the need for teachers and students to consider practical work as a thinking activity in which each participant constructed understandings, rather than solely the domain of the manipulative work of the hands. This was in
More pedagogies
91
sharp contrast to the predominant view of process skills promoted in the junior secondary science syllabus (Department of Education, 1978) and the developing primary science syllabus (Department of Education, 1979). This outcome of the in-depth phase was discussed further in the seminal article by Millar and Driver (1987), in which they critiqued the process approach to science education, argued that content and process cannot be separated, and promoted the development of thinking skills as a valuable curriculum and pedagogical aim. Sheema Saeed (1997) researched the perceptions of 12 middle and secondary teachers (from Malé, Maldives and Hamilton, New Zealand) of practical work, using interviews, a concept mapping exercise and a questionnaire. The teachers’ responses indicated that teachers’ perceptions of the term ‘practical work’ varied from perceptions of all activities in which students were actively involved to perceptions of practical work in school science as laboratory work using chemicals and equipment. It was found that teachers’ perceptions of the term ‘practical work’ matched the purposes for which they used practical work and what they mentioned they did for practical work in the classroom. Maldives teachers said they used practical work, including demonstrations, for the enhancement of conceptual understanding. New Zealand teachers reported that they frequently used practical work, especially structured student activities and open investigations, for developing practical skills as well as for conceptual understanding. Mavis Haigh (1998) researched the use of partially open investigative work with senior (Year 12; aged 16 years) biology students in New Zealand. Using social constructivist and co-constructivist theorising, she researched the initial use of partially open investigative practical work, by senior biology teachers, as required by the then new 1994 senior biology curriculum (Ministry of Education, 1994b). The research questions were related to the benefit and constraints in using partially open investigations, the teacher support required and the ways in which the students’ abilities in carrying out this type of investigation could be enhanced. The use of partially open investigative work was incorporated into six investigations: •
The ecology section of the curriculum: –
•
Green streams; the effect of fertiliser in a stream
The cell form and function section: –
– –
Factor X: finding the two best sources of factor X, in order, from four given plant materials. Factor X reacts with hydrogen peroxide to produce a foam. Factor X is an enzyme that catalyses the breakdown of hydrogen peroxide Potatoes for dinner: the effect of potatoes being placed in water and osmosis Sweet export: selection, from five different varieties of apples, the two that contain the highest sugar content
92
More pedagogies
•
The plant form and function section: – –
Plant cells at work: factors affecting the rate of production of a gas given off in light conditions by single-celled plants Plants for dry conditions: the selection of plants for planting a dry bank on an industrial site
The research evaluated the investigations in three phases. The first two phases were conducted in a large, urban, co-educational, state secondary school. All four biology teachers and 98 students participated in the first phase (1993) (Haigh, 1993), and one teacher and 32 students continued to be involved in 1994 and the second phase of the research (Haigh, 2000). In the third phase (1995), 22 teachers from other secondary schools from around New Zealand participated in trialling the teaching intervention. Overall, more than 400 students participated in the study. The self-report data generated by students and teachers, in classroom observations, field notes, researcher’s diary, interviews with teachers and students, questionnaires, student work sheets, and student and teacher evaluation sheets of the investigations, were analysed within an interpretivist paradigm. Pre- and post-investigation self-report data were collected from students regarding their declared confidence at carrying out investigative work, self-evaluations of their handling of investigations and of their perceived learning. The findings indicated that the initial use of partially open investigative work was exacting of both students and teachers. The students required deliberate and focused teaching if they were to progress their scientific inquiry skills (Haigh, 2001). The students valued the opportunity to have more control over the direction of their practical work. They found investigative work motivating and claimed that it increased their learning. A noticeable finding was the multiplicity of learning outcomes, covering a range of understanding of context, content and procedure (Haigh and Hubbard, 1997). Some students accepted the opportunity for personal involvement, self-direction and responsibility very quickly, whilst others needed ongoing support and encouragement. The teachers acknowledged a cognitive and affective value for investigative practical work in a biology programme but reported considerable difficulties with the introduction and assessment of such a programme. A teachers’ guide has been produced (Haigh, 1995b).
Inclusive pedagogies During the 1980s and 1990s, curriculum development in New Zealand was promoting the notion of science for all students – those going on to be scientists, and in particular girls; Maori and Pacific students, who were then currently under-represented in senior secondary and tertiary science classes; as well as those who did not continue with further studies but who would be living in our scientific and technological society.
More pedagogies
93
The work of the Pacific students at Waikato has contributed to the work on the inclusion of Pacific students in senior science classes. Mesake Rawaikela (1995), in his masters research, investigated some of the factors contributing to the academic underachievement of native-Fijian students when compared to Indo-Fijian students. Interviews were conducted with 20 native-Fijian science students and six science lecturers, together with questionnaires from 80 first-year University of the South Pacific science students. A key finding was that there seemed to be a mismatch between the native-Fijian students’ cultural norms and those expected by the university, as well as low socio-economic status. One of the implications of the findings was that further research on the teaching and learning styles in native-Fijian students would be a way forward. Lorraine Evening (1998) undertook an exploratory study in Fiji to investigate the issues associated with the under-representation of girls in senior physics in Fiji, from within the framework of the girls’ decision-making processes. Eighty-nine Year 13 girls (44 physics and 45 non-physics) from eight secondary schools in Fiji participated in small-group interviews which explored, first, their perceived influences on their subject choices, second, their perceptions of the under-representation of girls in physics and third, their views about the nature of physics. In addition, national enrolment data from the Ministry of Education was obtained and analysed in order to ascertain the extent to which girls were under-represented in physics in Fiji. The thesis rejected arguments which appeal to biological or ‘natural’ differences in girls and boys to explain the lack of participation of girls in science, and physics in particular. Instead it argued that social explanations were more useful and that the researcher’s local contextual knowledge and experiences brought a valuable perspective to this study, enabling an analysis of the complex sociocultural issues involved from a perspective that has previously been missing from the international literature. The thesis also argued that the issues or factors associated with the under-representation are complex and closely interrelated, and that significant ethnic variations in responses and further variations within the major ethnic group of girls indicate the non-homogeneity of girls in Fiji. The under-representation and underachievement of girls in science, especially in senior science, was also researched. Jane Gilbert’s doctoral thesis (1997, 2001) examined the ‘problem’ of girls and science education using ideas drawn from poststructuralist theory. She argued that: •
Through the particular conceptions of knowledge and subjectivity on which science is based, it is not possible to be positioned simultaneously as a woman, and as the subject of scientific knowledge (or as a fully-fledged modern liberal individual). The discourse of science has an assumption of sexual difference; an assumption which not only remains unacknowledged but which is actively concealed. The meaning of the category ‘girl’ is constituted in particular ways with respect to the category ‘science’. Hence, through the particular assumptions of subjectivity and knowledge that are relied
94
•
•
More pedagogies
upon in the production of the discourse of science, it becomes impossible for women to position themselves as both the subjects of scientific knowledge and (embodied) women. Sexual difference is a necessary condition of possibility for the development of modern forms of rationality and citizenship, and for the development of modern science. The deconstruction of the conventional understandings and theoretical underpinnings of the sex/gender, mind/body, man/woman binaries is necessary to undermine the ability of these binaries to continue to drive the way in which we understand the relationship between gender and science. The purpose in doing this is to begin the process of unsettling the assumptions through which the problem of girls and science is currently produced, and to provide a basis on which the ‘problem’ might be ‘thought differently’ . . . (and) . . . to make explicit, and to problematise some of the (unacknowledged) assumptions which underlie the ways in which science and science education are widely understood. No progress with respect to the ‘problem’ of gender and science can be made until the assumptions which underlie the categories sex/gender/woman and science/science education are disrupted. Following Valerie Walkerdine, Jane Gilbert argued that what is necessary is not to continue to provide empirical evidence that women are capable of participating and achieving in science, but to look underneath the surface for the assumptions through which the ‘truth’ of women’s lack of capability is constituted. A new model of sex, gender and embodiment can be developed which is capable of exceeding (rather than replicating) the binaries through which women’s exclusion from science is produced. This development of new forms of science education will encourage students to develop a critical understanding of scientific knowledge, to develop the ability to both ‘use and refuse’ (Fuss, 1989) knowledge that has been developed within science, and will allow young women to position themselves both as women and as the subjects of scientific knowledge. This new model for thinking about sexual difference, while acknowledging the existence of the gender difference, attempts to disrupt its conflation with biology and with the category ‘woman’ (and the resultant assumption of it as immutable). The purpose of this model is to open up new possibilities for thinking about sexual difference in ways which do not simply replicate the mind/body split and/or the ‘A’/‘not-A’ patterns which were described above. The model has two strands. The first, while distancing itself from the conventional conceptions of sexual difference and the category ‘woman’, involves the recognition of the existence of sexual difference and of women’s specificity. This is a position that, because it originates in the theoretical frameworks which are very different from the egalitarian/ liberal, social constructionist and/or radical feminist positions which have, until recently, been the basis of the vast majority of work in this area, opens up spaces for the development of very different kinds of
More pedagogies
95
political action. The second strand is political. It involves the development of strategies for action-in-the-world (in this case, the world of education); strategies which arise out of this rethinking of sex /gender, and which will produce changes in the Symbolic Order, and, as a result, in the ways in which our thinking is organised. (p. 313)
•
The model outlined is based on the view that it is more productive to find ways to develop an awareness of the processes through which woman has become the ‘other’ of science, and then, through the disruption of the knowledge claims which rest on this ‘othering’, to provide spaces within which it will be possible to conceptualise women’s specificity. Curriculum materials can be produced which invite students to rethink sex and gender; to see it, not as fixed property of individual bodies, but as a performance: to see it not as a given, but as something which can be used strategically in a variety of ways . . . driving all of this, however, is the contention that, in refusing the traditional construction of ‘woman’ as whatever is ‘not-man’ [i.e. woman are deficient], we will be able to create spaces in the Symbolic realm within which it will be possible for ‘woman’ (and female authority) to develop as autonomous categories. (p. 353) Alternatives to the teaching of the topics of animal behaviour and human evolution were outlined: an approach which, I argue, will make it possible for students to understand these conceptions in the way in which biologists understand them, but to, in addition, go beyond these understandings to develop a critical, deconstructive relationship with them. (pp. 473–4)
In summary, Gilbert argued that, through their reliance on modernist understandings of the categories ‘girl’/‘woman’, ‘sex’/‘gender’, and ‘science’/‘education’, previous approaches to this problem have reproduced – rather than solved – it. Following from this, a ‘deconstructive’ approach to each of the categories listed above was mapped out. Suggestions were made as to how this approach might be used to inform the development of science education programmes which can allow young women to conceptualise themselves as women and as the creators of new scientific knowledge. Another aspect of inclusiveness is that of including students who are gifted. Ian Francis (1998), in his masters research, investigated the conditions in which students considered gifted in physics find themselves in the New Zealand educational setting. A survey was sent to all 377 New Zealand secondary schools to ascertain their policy with regards Students with Special Abilities (SWSA), with
96
More pedagogies
143 (38 per cent) being returned. Responses were largely positive but there was little evidence of intentional, systematic addressing of the needs of SWSA in any of the respondent schools. Interviews were also undertaken with 12 male students, identified as gifted in Physics, from five Waikato schools, as to how they would like to learn physics. The findings indicated that: • • • • •
there is not a stereotyped SWSA student for whom a generic programme could be designed and directed at; they did most of their learning of physics in the classroom; they expressed a wish for a challenge; the students had little understanding of their own learning process; they felt constrained by the system and would prefer broader studies, university courses taught by video link, etc., university courses taught in the high school, honours classes, extra research and practical investigation, peer teaching, concurrent university enrolment, grouping by ability, mentoring, e.g. using electronic means, and tiered assignments.
Metacognition An aspect of teaching that takes into account students’ thinking is students thinking about their thinking: the metacognitive. Students who are more metacognitive about their own learning have better learning outcomes (Baird, 1984; Baird and Mitchell, 1986; White and Gunstone, 1989). At Waikato, addressing metacognition, as a part of teaching that takes into account students’ thinking, has been in the research associated with learning dispositions (Carr, 1997; Carr, 2000; Hameed, 1997) in both science and technology education. Hassan Hameed (1997) reported on his doctoral thesis which explored critical thinking dispositions in the context of secondary-level students in the Maldives. Previous attempts at teaching ‘thinking’ in science had been invariably restricted to the development of thinking abilities. New theories of learning, such as social constructivism, argue that abilities-based instructional efforts are too narrow because human behaviour is determined by both abilities and tendencies such as inclinations and sensitivities. A recent conceptualisation of a theory of thinking which includes these components is the dispositional theory of thinking (Tishman et al., 1993, 1995). This theory postulates that seven dispositions, each comprising abilities, inclinations and sensitivities specific to the disposition, are necessary for good thinking. These dispositions are: • • • • • • •
to be broad and adventurous; toward sustained intellectual curiosity; to clarify and seek understanding; to be planful and strategic; to be intellectually careful; to seek and evaluate reasons; to be metacognitive.
More pedagogies
97
Perkins, Jay and Tishman (1993) contend that dispositions are acquired through social interactions and enhancing thinking can be fruitfully viewed as enculturation into a community of practice. Students are immersed in this community of practice, and exposed to social practices (language, models and interactions) that emphasise thinking. In the first part of the study, in 1994, the prevalence of thinking dispositions of 100 secondary-level students in the Maldives was assessed by a survey. Data from the survey were triangulated by interviewing 10 science teachers. These data indicated that the following four dispositions were least prevalent in the sample: the dispositions to be broad-minded, to seek clarification, to be metacognitve and to be planful. Furthermore, no gender disparities in the dispositions could be identified. Based on a model of teaching referred to as thinking apprenticeship, an intervention was developed to enhance the least four prevalent dispositions. In the intervention, the thinking instruction was incorporated into the physics content of light and taught within the physics periods. Based on the suggestions of Tishman et al. (1993), cultural exemplars, cultural interactions and direct instruction in cultural knowledge were used to enhance dispositions. The physics content was used as a vehicle for enculturating the thinking dispositions. The strategies used to enculturate dispositions were: modelling and demonstrating; articulating and coaching, scaffolding; and reflection and feedback, and exploration (that is, the deliberate transfer of learned dispositions to other contexts). The intervention was trialled, in 1995, in two secondary schools in the Maldives with over 60 students using pre-test/post-tests of the intervention group and control group (matched on scholastic ability) for about two months. The findings indicated that there were significant differences in students’ thinking dispositions between the intervention and control groups following the intervention. The intervention group improved their dispositional thinking compared to their own initial level and compared to their counterparts on the control group. Abilities were found to be sensitive to intervention. Enhanced performances were observed not only in the physics area but also in non-physics everyday topics, suggesting transfer of thinking across domains. It was argued that this evidence supports the interpretation that the students’ improved achievement was caused by enhanced dispositional thinking and not by improved domainspecific skills. The findings also substantiated the claim that infused approaches based on thinking apprenticeships are compatible with contemporary schooling systems and hold good promise for making learning more meaningful. Ian Taylor (2000; Taylor, Barker and Jones, 2003) developed and trialled a model-building approach designed to teach astronomy to New Zealand Year 7 and 8 students. Although Science in the New Zealand Curriculum (Ministry of Education, 1993b) emphasises the integration between the learning of investigative skills and the learning of astronomy ‘content’, the teaching of astronomy appears to remain fact-based and transmissive in style (Taylor, 2000). The difficulties that learners may have in using and understanding mental models for astronomy concepts are well documented in the research literature (Pfundt and Duit, 1994),
98
More pedagogies
suggesting that transmissive pedagogy is not likely to be successful in helping many students meet the intended ‘content’ outcomes. With this challenge in mind, Taylor’s approach drew on Hesse’s (1966) suggestion that the efficacy of a model can be judged by a process of learners (rather than the teacher) repeatedly critiquing the model in terms of its positive, negative and neutral attributes. Positive attributes are those properties which both the model and the mental model share and negative ones are those attributes not common to both the model and the mental model. Neutral attributes are simply those attributes not yet classified as positive or negative (Hesse, 1966). Taylor’s intervention comprised 11 lessons that proceeded according to a four-phase sequential teaching structure: 1
2
3
4
The Focus on the Mental Model Phase (Lessons 1 and 2) explored ideas about mental models and actual models. The technique of critiquing models to show that all models have limitations was introduced during this stage and the pupils’ mental models were elicited and shared. The Model-Building and Critiquing Phase (Lessons 3 to 7) focused on the use of scientists’ mental models for the solar system to construct an actual model (called an orrery). The actual model was a commercial version of an orrery and the students initially expected that, because it was manufactured, it would be an accurate portrayal of the reality. The class repeatedly critiqued the ability of the orrery to illustrate the scientists’ concepts, with the intention of helping them understand the intelligibility and plausibility of the scientists’ mental model. Each lesson in this phase included a focus– challenge–review cycle of the concept under investigation in that lesson. The Using the Scientists’ Mental Model to Solve Problems Phase (Lessons 8 and 9) encouraged pupils to utilise the scientists’ mental model to solve some problems which were novel to them. Thus the focus of their thinking shifted from intelligibility and plausibility to fruitfulness. Novel (for pupils) problems that were explored included the causes of tides on Earth and what might happen to these tides if Earth had two moons. Each of the problems was resolved by several groups of pupils, thereby ensuring an attentive audience for the final reflective phase. The Reflection Phase (Lessons 10 and 11) focused on reporting and debating solutions of the different groups to each of the problems posed. This was intended to help the pupils further consolidate the scientists’ mental model.
The pupils’ astronomy knowledge was assessed in a pre-test, and the results were compared with tests of astronomy knowledge immediately following, and four weeks after, the intervention. The results showed that pupils learned many astronomy concepts as a result of the intervention and that this knowledge was retained for over four weeks. Taylor interviewed pupils and found that they responded positively to the learning–teaching approach adopted in the astronomy intervention, showing a clear affinity with the idea that they could attempt, as
More pedagogies
99
a group, to resolve problems in their own way within an overall supportive environment. They appeared to find the cooperative problem-solving and reporting-back phases to be interesting, challenging and informative, particularly when they realised there were conflicting views that needed to be resolved.
Classroom discussion Another theme that emerged from the LISP research programme was the promotion of classroom communication as a worthwhile pedagogy in science education. The five Learning in Science Projects and related theses have researched effective communication in classrooms. The main focus of this research has been on pedagogies to promote classroom interactions and communication (teacher–student, and student–student) so as to promote conceptual development. However, the argument has remained that of the pivotal role of dialogue and communication (both oral and written) in promoting conceptual development. The Learning in Science Projects and associated theses research on pedagogies based on the role of language and communication in the learning of science is documented in the following sections. Meanings for words In the first Learning in Science Project, the role of language was discussed in the context of the mismatch between the students’, teachers’ and the accepted scientific concepts (Bell and Freyberg, 1985). The problems in communication were viewed as different meanings being constructed by participants in the classroom either in teacher–student or student–student communications. This was not only due to some of the technical words of science being unfamiliar to students (and some teachers) but also because even simple words can have different meanings in differing contexts. For example, the word ‘animal’ is an everyday word that is used differently in an everyday context (a four-legged furry creature), and a scientific context (a consumer). Calling a person an animal in an everyday context has a different meaning to their being called an animal in a scientific context. To help students differentiate these differences in meanings, small discussion groups were used to enable student–student discussions as well as teacher– student discussions. For example, the card game based on students sorting interview-about-instances cards was trialled (Bell and Freyberg, 1985). The students had to reach a consensus about the categorisation of an instance before it was placed on one of two piles: the animal pile and the not-animal pile. Other language-based activities trialled were crosswords and dice games. Mindful that teachers have a responsibility to teach the scientific concepts, as legislated in the national curriculum, the caution was made that these discussions do not always result in the scientifically accepted concept being constructed by students. The scientists’ concept may have to be provided by the teacher or a student. But this does not have to be done by the giving of a lecture.
100
More pedagogies
Ways of communicating the scientists’ ideas Lynne Oxenham (1995) documented the 29 ways (including telling the students) in which four secondary teachers introduced the scientist’s ideas into the classroom dialogue. The research was done in response to teachers who take into account students’ thinking, often categorised as student-centred teachers, with an oft-expressed view that these teachers do not ‘teach the science’. But what is meant by ‘to teach science’? This masters thesis investigated how some teachers introduced the science ideas to their students as a part of their teaching. Four teachers of science, and a total of 28 students ( Years 9–12) taught by them, from a single co-educational secondary school, were involved in the research. Data were collected from 16 lessons using classroom observations and semi-structured interviews with the teachers and students. The findings indicated that the dialogue about science ideas was identified as occurring in two stages: an introduction stage and a post-introduction stage. Within the introduction stage, 18 ways of introducing the science ideas were identified. In the post-introduction stage, six categories and 11 themes were identified. These were: •
The introduction stage 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 17 18
•
Teacher directly tells the science idea Teacher retells the science idea Using a diagram Using a model Using personal experience Questions calling on students’ current understandings Teacher questions using students’ own words/questions Via teacher cues Discussing the purpose of experiments Using experiments and discussing results Via student dialogue Via student ideas Via student questions Direct recall questions of previous lesson ideas Teacher correction of student ideas – from ‘material’ Direct response to student question Revisiting ideas – via student introduction
The post-introduction stage 1
2
Confirmation – Dialogue theme 1: Confirmation of student answers to teacher questions – Dialogue theme 2: Confirmation of student idea Expansion – Dialogue theme 3: Expansion of an idea via a teacher’s question – Dialogue theme 4: Expansion of an idea via a different example
More pedagogies
– 3
–
Dialogue theme 8: Linking of ideas from previous work Dialogue theme 9: Linking ideas from the current lesson or discussion
Teacher requests for justification of student ideas –
6
Dialogue theme 6: Clarification via student in response to teacher question Dialogue theme 7: Clarification via another example
Linking ideas – –
5
Dialogue theme 5: Expansion of an idea or answer provided by a student
Clarification –
4
101
Dialogue theme 10: Teacher requests for justification of student ideas
Summary of ideas by the teacher –
Dialogue theme 11: Summary of ideas by the teacher
Hence, the teachers employed multiple ways to introduce the science ideas to the students, indicating the multiplicity of communication types used by these teachers of science. Reading texts as a pedagogy to learn science Another form of communication in the classroom is the reading of texts, although there has been little done at Waikato following the interest in readingto-learn in the science education in the 1970s and 1980s. Beverley Bell (1984a) researched reading as a pedagogy for conceptual development in science education, linking the conceptual development research with that of reading comprehension. She investigated the role of students’ existing knowledge (of ‘animal’) in reading to learn science, that is, in both the reading comprehension and learning processes. Three different methodologies were used – one in each of the three phases of the research. These were the qualitative interview techniques (‘spot-the-mistake interviews’, n=six Year 9 students and the readingto learn-interviews, n=21 Year 9 students) and quantitative survey measures before and after a class reading task, n=227 Year 9 students. The findings indicated that the students’ existing knowledge not only contributed to the meanings constructed whilst a student was reading a text to learn science, but it also influenced what constructions were made. The existing knowledge was also used to evaluate the constructed meaning, in terms of whether to accept or reject the construction, its status. Hence, the careful explication of the scientists’ meaning in a text by the author does not guarantee that the reader will construct and accept the intended meaning or learn it.
102
More pedagogies
The use of narrative as a pedagogical tool Whilst the Learning in Science Projects have used, to varying degrees, the narrative as a research tool, it has also been investigated as a pedagogical tool. For example, in the Learning in Science Project (Teacher Development), the preferred learning activity of the teachers was that of sharing anecdotes (Bell, 1994c). The teachers most valued talking and listening to other teachers. Most teachers saw it as an important way to learn and develop. One way of talking with the other teachers was by using anecdotes (a narrative of a significant event). ‘Telling anecdotes is an everyday way to make sense of our experiences to ourselves or to add sense to what has happened to us. In telling an anecdote, a teacher can talk of experiences and actions and become aware of the beliefs, assumptions and feelings underlying them’ (Bell and Gilbert, 1996: 122). Hence, it became the main pedagogical tool to promote the learning by the teachers. ‘The anecdoting and accounting provided the interaction necessary for cognitive development and enabled the tacit knowledge, values, norms and morals of the teachers to be discussed, renegotiated and reconstructed’ (Bell and Gilbert, 1996: 122). It also gave support to the social nature of cognition. In similar ways, students in classroom discussions have shared anecdotes based on their past experiences. For example, Rodrigues (1993b) gives many instances of when the teaching and learning activities promoted the sharing of anecdotes. In the unit of work ‘Driving teenagers to alcohol’, the students shared experiences of themselves and others having been breathalysed by police in random breathchecks. They also shared experiences of hair perming, and farm activities. The anecdoting was a way of linking prior experiences to the science being learnt. In other words, the anecdoting was a part of the discussion that resulted in conceptual development. Classroom discussions Discussions by small groups of students and discussions by the whole class have been a feature of all five Learning in Science Projects and related thesis research at the University of Waikato. Their use as a pedagogical strategy to promote conceptual learning is argued on the ground that specific focused discussions avoid the typical patterns of teacher–student interaction and enable the students to use skills and competencies that promote conceptual development. The smallgroup and/or whole-class discussions provide the opportunity for the students to clarify and share their own understandings; to compare their own conceptions with those of others; to test out their understandings; to ask questions and to challenge the views of other students and the teacher; to reconstruct their understandings and to use the new ideas with confidence. The effectiveness of discussion in small groups in promoting conceptual development was evaluated positively in the theses already discussed: Barker, 1986; Happs, 1984; Kirkwood, 1988; Rodrigues, 1993b; Stead, 1980a; Taylor,
More pedagogies
103
2000. The discussions usually emerged during a specific and planned activity such as card sorting (Stead, 1980a), crosswords (Bell, 1981a), task sheets (Stead, 1980a), practical work (Tasker and Freyberg, 1985), or rock sorting (Happs, 1984). It was not the activity per se that promoted the learning but the language and thinking involved in completing the activity. This widening of the notion of small-group discussion work beyond that involved in doing practical work was a feature of 1980s curriculum development (Ministry of Education, 1990a) and the 1993 national science curriculum development (Ministry of Education, 1993b). In an additional study, Mala Raghavan (1988) investigated the effectiveness of a peer interaction study to promote conceptual change in students’ views of ‘living’. Peer groups of two students engaged in a booklet task which allowed for the discussion of the students’ own ideas, those of their peers and the scientists’ view. Surveys and interviews were the main data collection techniques used. The main findings were that the peer interaction strategy used in this investigation appeared to enhance immediate conceptual change. Conflict, either with pupils’ pre-existing ideas, their peers’ or the scientists’ science, seemed to be an important factor for change, whilst individual interaction with material, with the researcher present, facilitated change in some cases. However, the peer interaction may sometimes facilitate change towards unintended outcomes, that is, towards children’s science or mixed (school science and children’s science) views. In another study, Wheijen Chang (2000; Chang and Bell, 2002), in her doctoral work, researched a three-week teaching intervention of classroom discussion, taught by the researcher as part of her usual work as a physics lecturer, of first year tertiary physical science and engineering students in Taiwan. Each year, the university has 2000 enrolments in this prerequisite course for all science and engineering students. The teaching intervention was based on constructivist and sociocultural views of learning and was designed to address the problems of lack of interest and attention of students in the physics classes, the ‘embarrassing dozing problem’, the silence and lack of engagement in class, and the deterioration of attitudes towards learning physics throughout the course period and low achievement rates of the students. The intervention teaching promoted the students’ participation in the classroom through individual work on conceptual questions, small-group discussion and whole-class discussion. The course content was also changed so that the physics was taught through phenomena and verbal descriptions; using more life examples; reducing the number of physics problems (in the textbook) to be solved; providing more challenge; and ensuring the teaching was different from that which the students had already had in high school physics. One of the research questions was: what are the learning outcomes of the intervention teaching in comparison with the traditional teaching? The research was in three phases. In Phase 1 (1997), the 206 first year students were surveyed on their entering learning attitudes and expectations of the university physics course. 256 second year students who had completed the university course were also
104
More pedagogies
surveyed for their perceptions and comments on the physics course. In Phase 2 (April 1998), 330 students were surveyed about the course and eight physics lecturers interviewed about their views on teaching and learning. In Phase 3 (September and October 1998), the teaching intervention was implemented and evaluated. The intervention teaching was both quantitatively and qualitatively evaluated using pre- and post-tests of academic conceptual understanding, pre- and post-questionnaire surveys and interviews with students on learning attitudes and perceptions of the course. The lecturers’ opinions of the course were also investigated. The evaluation of the intervention class was also compared to a similar evaluation of a traditionally taught class, taught by another lecturer. The evaluation of the intervention and traditional teaching showed that • •
•
• • •
the students in the intervention class had improved perceptions of interest, satisfaction, enjoyment and achievement; the intervention teaching also promoted students’ engagement in learning physics in class, including their willingness to attend, listening to the lecture, and participating in discussion; the programme improved their learning commitments out of class, including study hours and the adoption of a comprehensive, deeper-level learning approach. ‘Therefore, this program “cured” the students’ dozing as well as the lecturers’ sore throat at the same time’ (p. 322); while the programme enriched the students’ perceptions of learning by means of providing more challenge, it had not undermined their confidence; the intervention class students expressed their appreciation of the new teaching approach; the percentage of high-risk students, i.e. students who are likely to be failed in their final grades, was reduced.
However, the students in the intervention class were found to have a similar academic achievement in physics to that of their peers in the traditional classes: With the significant improvement in both attitudes with respect to physics classes and perceptions of learning physics, an improvement in academic performance might be anticipated if the programme was extended over a longer period. Meanwhile, the traditional assessment design and techniques are perhaps problematic in terms of not assessing the goals of the course. (p. 323) Findings on the dialogue in the classroom were also obtained from the Learning in Science Project on teacher development, as the teachers had indicated that an area of professional development they would welcome was that of developing their skills of interacting with students to discuss and explain science concepts. In taking into account students’ thinking in their teaching, these teachers
More pedagogies
105
were responding to and interacting with the students’ thinking that they had elicited in the classroom. A central part of this teaching is dialogue (not a monologue) with students to clarify their existing ideas and to help them construct the scientifically accepted ideas (Scott, 1999). Therefore, giving feedback to students about how their existing conceptions relate to the scientifically accepted ones, and helping them to modify their thinking accordingly, is a part of teaching for conceptual development. It is also a part of formative assessment and formative assessment is seen as a crucial component in teaching for conceptual development (Bell, 1995a; Cowie, 1998). The teachers were therefore undertaking formative assessment whilst teaching for conceptual development. The role of meaningful contexts to promote effective discussions is acknowledged. Rodrigues and Bell (1995) analysed the interaction of some discussions by a whole class and in small groups, as part of a larger study (Rodrigues, 1993b). The discussions took place within a programme of work based on contexts that were meaningful for Year 12 chemistry students in a unit of work on oxidation and reduction. The contexts included Thiobaccillus bacteria and the New Zealand farmer, hair perming, breath testing, and aluminium production. The units were used over a period of 10 lessons in a class of 20 Year 12 female students. Hence, the talking, which was analysed, occurred when students’ own existing knowledge was accessed and when the students were able to make links to it. The talking occurred during a variety of pedagogical approaches and activities, including role plays, class discussions, experiments and group problem solving. The findings indicated that the typical teacher initiate–student respond– teacher evaluate and initiate–respond–follow up patterns were absent. Instead, the classroom discussion was characterised by the students and teacher collaborating and sharing their expertise. The students increased their understanding whilst mediating propositions, issuing assertive questions, volunteering information and sharing expertise with the teacher. There was some measure of gain in the students’ redox conceptual understanding as elicited before and after the teaching using pre- and post-questionnaires, concept mapping and formal school tests. In addition to their declarative knowledge, the students also developed a variety of cognitive skills including: • • • • •
constructing questions, proposing new questions, setting up hypotheses in the form of questions; volunteering evidence drawn from other sources, therefore making links to other experiences; communicating through recreated experiences; being reflective; challenging each other and the teacher. There was a notable shift in expertise from the teacher to expertise shared with the student.
It is proposed that these cognitive and social skills are the ones used in the conceptual development process.
106
More pedagogies
A focus on the social, rather than the cognitive, aspects of classroom interactions was a feature of the study (Wigglesworth, 1999). Of special concern were the social ‘barriers to learning’ for the ‘at risk’ students and students with low self-esteem, the main research question being ‘what aspects of the classroom social environment significantly influenced the construction of scientific understandings?’ The research involved a five-week teaching module on photosynthesis, taught to a class of 14-year-old ( Year 10) students. The students were asked directly to name their understanding of what events/situations working with others within their class helped or hindered their learning of science concepts. Their personal reactions to some of these social interactions were also obtained. Self-report data were obtained by student and teacher/researcher observational and interpretive daily journals and also by way of interviews and open-ended class discussions between the teacher/researcher and the students. The classroom aspects that the students said helped their learning were • • • • •
practical activities; class and small discussion group discussions; teacher and student question/answer sessions; teacher explanation of the science concepts; written work, in that it offered an opportunity for students to review and reflect on their work, particularly their homework time.
The classroom aspects, which were viewed as hindering learning, and which were consistently listed by students, included: • • •
affect – largely negative feelings triggered within the learner as a result of some social event/situation that generally occurred within the classroom; noise – if there was more than a low level of ‘working’ sounds within the classroom; a rushed pace – if the speed at which the teacher moved through the concepts involved in the topic being taught was perceived by the students as ‘too fast’ for them to ‘grasp’ the conceptual understandings.
Affect was seen to play a role in learning in the science classroom. Negative feelings were perceived as having a more powerful impact and were shown to interfere with the students’ ability to concentrate on their learning of the science concepts, so hindering learning. Negative affect particularly came into play when students felt their learning goals were being thwarted by some event/situation within the classroom or when an interpersonal relationship became somewhat ‘dysfunctional’. (Wigglesworth, 1999: 206)
More pedagogies
107
It is these affective responses that impact on the learning of science. Discussions and interviews were also explored as a more valid assessment tool (than written responses) for Maori students of science (Kent, 1996). This research is more fully discussed in Chapter 8. Viewing interactions for learning as sociocultural practices was researched in a tertiary science education context. Chris Eames (2000, 2003a, b) investigated the learning of tertiary science and technology students whilst on work placements as part of a cooperative education (or sandwich) degree programme at the University of Waikato. Eames used an interpretive methodology to collect the data. Postal surveys were received back from 95 of 125 recent graduates of the programme. The surveys contained a mix of open, qualitative and closed, quantitative questions. The results of this survey laid a foundation for the interview questions in the subsequent, longitudinal study. The 22 second year students accepted an open invitation to be involved in the longitudinal study in which they were interviewed on at least four occasions over 18 months and mainly whilst on work placement. These students were also encouraged to keep a journal or diary. The findings indicated that learning was achieved by working alongside a practising professional, engaged in authentic practices, as the students gained legitimate access to a new community of practice and became enculturated into ways of thinking and acting as a member of that community. The data analysis suggested theorising of the workplace learning as a sociocultural practice, involving mediated, situated and participatory practices. The students learnt through social interactions, and the use of the language and other tools of the community. The findings indicated that the students learnt about the practice of science and technology through their participation in science and technology workplaces. They came to understand the nature of the workplace enterprise, and developed specific skills and knowledge that allowed them to feel a part of their community. This study found that student learning in cooperative education placements can be seen as complementary to university classroom learning. Students felt they could integrate their learning between the two settings when they could apply discrete skills or knowledge from one to the other. Yet, for some students who could see little integration between the two settings, learning in each setting may still complement the other to provide a valuable education. The students noted different learning and assessment modes between the two environments, indicating that they would benefit from a pedagogical and curriculum design in their cooperative education programme that would help ease the transition between the two. From a sociocultural perspective, these cooperative education students can be seen as learning within two distinct socially and culturally determined communities of practice – the university and the workplace. How they are able to make connections between these communities, with the help of the cooperative education placement coordinator, will determine how they make the transition from student to science and technology practitioner.
108
More pedagogies
Teacher education and development for new pedagogies All the research on new pedagogies in science education undertaken by the Waikato group, has indicated that teacher development is typically required before the new pedagogical approaches can be implemented. Using the new teaching approaches is not simply about using new teaching and learning activities in the classroom; many other aspects of pedagogy needed to be changed or developed, for example the teachers’ views of teachers, teaching, learner and learning, and the nature of science. The fourth Learning in Science Project ( Teacher Development) researched the ways teachers might be helped to develop their teaching. The findings of the project and associated theses are discussed in detail in Chapter 9. Here, the key points relating to promoting pedagogical change are summarised. The first point to be made was that teacher development for pedagogical change needed to have a focus on the teacher and teaching, not the learner. When the first Learning in Science Project began, in 1979, much of the research in science teaching was focused on the teacher. The work of the Waikato group contributed to a shift in focus to the learner, and this was reflected in the communication of the findings to teachers. This focus on the learner was powerful in helping teachers become aware of the need for change and promoting a change in their beliefs and professional knowledge. However, to promote a change in teacher practice, the highlighting of teaching and teachers was needed. Initially, the teachers in the Learning in Science Project (Teacher Development) wanted to talk about what they were doing in the classroom and about themselves as teachers. The teachers had initiated the change in the classroom, not the students, and therefore wanted to focus on the changes they were making and how they felt about the changes. This initial focus on teaching (and indirectly on learning and students) was a factor that helped the teacher development (Bell, 1993c). Another point in the findings was the ‘feeling better about myself as a teacher’. The teachers commented that using the new teaching activities had helped them to ‘feel better about themselves as teachers’ or to be ‘more like the kind of teacher I would like to be’. This was felt to be a pay-off to continue to change and grow professionally despite the difficulties that the change process involved. In other words, teacher development was enhanced as the teachers felt better about themselves as teachers as a result of making changes to their teaching activities. The teachers’ sense of self as a teacher was an aspect involved in the teacher development process. For some teachers, the new classroom activities enabled their actions to match their beliefs about what it means to be a teacher of science. For others, the programme helped them develop their beliefs about being a teacher. For example, many of the teachers changed their view of noisy classrooms. A noisier classroom may not necessarily indicate a lack of control – it could mean more discussion between students and improved learning. The
More pedagogies
109
teachers’ criteria for evaluating their teaching were changing. The teachers tended to feel ‘better about themselves’ because they • •
• • •
•
•
•
•
•
•
found value in listening to the students rather than just talking at them; knew their students better, were pleasantly surprised how much the students knew about different topics, and that they now had a way to communicate to the students the value they attached to what the students brought to the lesson. In many cases their respect for their students’ knowledge had increased; enjoyed being able to get insights into what was happening for the students as they were learning; felt good about being able to develop a greater sense of trust that the students would learn something; appreciated spending more time interacting with the students’ ideas rather than interacting with them for organisational or management reasons. The teachers reported less need for behaviour management; enjoyed having fewer demands being made directly on them by the students as the students were thinking more for themselves and not being so dependent on the teacher; appreciated that the responsibility and energy for keeping the lesson moving did not rest solely with them as the students were taking more initiatives in the lessons. The teachers felt they had to spend less time and energy motivating the students to learn; felt that the new teaching activities did not mean that they had to stop doing some things they they felt to be central to their view of what it means to be a teacher. They still felt that there was a role for their teaching activities of giving some structure to the unit of work, planning, covering the syllabus, student grouping in the lessons, facilitating class discussion, and assessment. They still felt that they had control over what was happening in the lessons and that they were doing those things required of them by the students, school, parents and the government; learnt alongside the students. They enjoyed being learners themselves and they appreciated the lessening of the expectation that they had to know all the answers; enjoyed the different relationships with the students. The teachers felt that they related and talked with the students in a different way. They felt the new teaching activities had enabled them to work more alongside the students rather than from the front. They felt good that the students were enjoying being in their lessons; felt good that the students were learning ‘better’. (Bell, 1993c)
‘Feeling better about myself as a teacher’ arose because the teachers had changed what they did in the classroom, especially the way they interacted with
110
More pedagogies
the students, and they had changed their ideas and feelings about what it meant to be teaching and to be a teacher. If the teachers had not made these changes, they would not have experienced what it felt like to be a ‘better’ teacher. If the new teaching activities had not helped the teachers to ‘feel better about themselves as teachers’, then the teacher development would not have continued. Another aspect that promoted pedagogical change was that of ‘better learning’. When talking about the classroom feedback they had received on the new teaching activities from students, the teachers initially talked about ‘better’ learning (Bell and Pearson, 1992a). That ‘better’ learning was occurring was perceived by the teachers as a reason to keep on using the new activities in the classroom. The teachers were comparing their perception of the learning occurring with former teaching to that occurring with the new teaching. The ‘better’ learning was seen as a payoff to keep on developing their teaching. The comments about ‘better’ learning included those related to better learning conditions (enjoyment, social cooperation, ownership, student confidence, motivation) and ‘better’ learning outcomes (responses to teacher questioning, debates and written work, the development of students’ ideas and the transfer of learning). Later in the programmes, the teachers continued to think about and discuss learning as an important aspect of the feedback they were receiving about their new teaching activities. However, the emphasis changed from commenting on the ‘better’ learning occurring with the new teaching activities to comments on indicators of learning, and from talking about learning conditions to learning outcomes. In summary, the factors that helped teacher development were: • • • • •
‘better’ learning with the new teaching activities compared to that with former teaching; being more aware of and seeking more information in the classroom on learning; focusing more on learning outcomes than learning conditions; a focus on teaching and what the teachers have to do; support, feedback and reflection.
May Cheng (2000, 2002a, b), in her doctoral thesis, investigated the learning of novice primary teachers of science, in a teacher education situation in Hong Kong, as they learnt how to teach using teaching approaches based on a constructivist view of learning. In this two year study, the researcher/teacher educator followed the teacher development of the same group of student teachers during a teacher education programme to the time after their graduation and after their first year of teaching. Thirty-nine students were initially involved, and all were taking science as their main curriculum area, with fewer being involved in different phases and data collections. The study was conducted in three phases. The first phase (September to January 1997) involved the implementation of the Curriculum Studies module
More pedagogies
111
in the teacher education programme. The module introduced a range of science teaching approaches in which there was an emphasis on teaching science to primary pupils, based on a constructivist view of learning. The student teachers were interviewed before and after the module, to ascertain whether there were any changes in the views of the student teachers on science teaching and learning. The teacher education sessions were recorded, field notes taken and questionnaires given to the student teachers to complete. During the second phase (March to May 1998), the student teachers were observed in the lessons they taught whilst on practicum; field notes being recorded; they were interviewed to find out whether the module had influenced their views of teaching and learning in the primary school context, and given a questionnaire to complete. The researcher/teacher educator also provided support group meetings. In the third phase (September 1998 to June 1999), the beginning teachers were interviewed three months into their first teaching employment. The interviews provided information on the kinds of support needed by these beginning teachers. Subsequently, a resource package was developed, and an introductory workshop and personal support from the researcher/teacher educator was provided. During the use of the classroom resource pack, the beginning teachers were observed teaching and were interviewed. A questionnaire and journals were also used for data collection. The two interventions – the Curriculum Studies module and the beginning teacher resource package (Cheng, 1999) and workshops – were specifically developed for researching in this study, and took into account the then current research and theorising on teacher education and science teaching practices. The Curriculum Studies module consisted of 12 sessions, each one lasting for two hours. Overall, both qualitative and quantitative data were collected. The surveys included a student teacher background questionnaire, the science teaching efficacy belief instrument, the science teaching questionnaire, the post-teaching practice questionnaire, and the beginning teacher questionnaire. The research data were collected separately from the assessment of the student teachers. The findings suggested that the student teachers experienced increased confidence in teaching science after the Curriculum Studies module in their teacher education programme, and were able to teach with a constructivist view of learning in their teaching practicum. However, as the graduates started their first teaching employment, they were often unable to teach in the ways learnt in the teacher education module. With the support of the resource pack and the researcher/teacher educator, the science teaching of the beginning teachers was more consistent with a constructivist view of learning and some were even able to influence their colleagues to rethink their science lessons. Hence, the two interventions – the Curriculum Studies module and the beginning teacher resource package and support – appeared to have facilitated the novice teachers to teach with intentions consistent with a constructivist view of learning; to use practices that engaged primary pupils in thinking in the lessons;
112
More pedagogies
to develop better learning conditions and outcomes; and to experience gains in their confidence in teaching science. The teacher education and teacher development practices were theorised as sociocultural practices.
Summary Many different teaching strategies, activities and pedagogies to improve the learning of science have been developed and evaluated in the Waikato research. The research evidence indicates that the pedagogies which have improved learning outcomes in the studies are: pedagogies that take into account student’s alternative conceptions; the teaching of science in meaningful contexts; a revised view of ‘practical work’; and classroom discussion. Many of the studies researched the inclusion of learning outcomes, which are additional to national curricula but are seen as valuable, such as dispositions. Collectively, the research has reported on multiple aspects of pedagogy, including the nature of the learner, the learning process, alternative conceptions, metacognition; views of mind, epistemologies, science student diversity and cultural aspects; science teacher knowledges, roles, professional skills, teaching activities, teacher development, assessment strategies, purposes and activities; science classroom discourse, values and norms in the science classroom; and science curriculum aims, purposes and desired outcomes, the science content to be taught and learnt, and progression in teaching and learning. However, what counts as evaluation is an ongoing discussion (Leach et al., 2003). In the Waikato research, the research methodologies used and data generated can be described collectively as interpretive, qualitative, quantitative, descriptive, collaborative, naturalistic, self-report data, triangulated, multiple data sources, classroom-based, pre-, post-, and post-post testing and guided by the ethics of care. Some of the data analyses gave valuable information for teachers as to what is happening during a particular teaching and learning situation, with surveys being used to find the extent of a finding. This qualitative and descriptive data had otherwise been hidden in much of the research prior to 1979, when the first LISP project began. This kind of research dominated the research at Waikato in the 1980s. Other research was evaluative, with quantitative data being collected to find out the learning outcomes; often the students’ learning being assessed against specified goals, including learning outcomes in the national science curriculum. Parallel classes were used for comparison in many of the research studies, instead of a ‘control’. Although this was not an experimental design as the many classroom factors cannot be controlled for, a comparison was possible, with ecological validity being maintained. The writing up of the research was done in a variety of ways to communicate to different audiences, such as teachers, teacher educators and curriculum developers, as well as Department/Ministry of Education staff. Hence, the data analyses have been done with an aim of communicating directly with the education
More pedagogies
113
personnel able to effect improvements in science pedagogy and learning outcomes, at the national and classroom levels. Over the twenty years of this research, the term ‘pedagogy’ has been used increasingly instead of ‘teaching’, as the term has a wider meaning (Bell, 2003). The term ‘pedagogy’ has not been a commonly used term in the AngloAmerican tradition of education. However, the term is becoming more widely used today and pedagogy has been more broadly described as including reference to the values, aims and philosophy of education, for example: ‘a method of teaching interpreted in its widest sense (Winch and Gingell, 1999: 170), including values, aims and epistemological considerations’ (Bell, 2003: 3). It is defined also with reference to the linking of power and knowledge, for example: the process of knowledge production . . . (including aspects of ) instruction and social vision . . . (and) political contexts . . . (with) concern for how and in whose interests knowledge is produced and reproduced. (Gore, 1993: 4–5) The situated nature of pedagogy is also acknowledged, for example: Pedagogy viewed from this situated perspective is not concerned with discrete teaching skills or techniques, but rather with the construction and practice of learning communities . . . a theory of pedagogy must encompass all the complex factors that influence the process of teaching and learning. (Leach and Moon, 1999a: 268) These definitions signal the complexity and multiple facets embodied in the current usage of the term ‘pedagogy’. To use the broader term of ‘pedagogy’ rather than the narrower term ‘teaching’ is to acknowledge: • • • • •
the discipline of teaching as a body of knowledge(s) that is systematically articulated; the physical, social, cultural, historical, economic situatedness of teaching practices; the interdependence of teaching, learning, knowledge, assessment, curriculum; the interrelatedness of pedagogy, culture, society, politics and economy; the power in the teacher–learner–curriculum relationship in the classroom.
Hence, pedagogy is viewed as being more than ‘best practice’, more than the techniques or strategies of arranging the seating in the classroom, choosing the materials and equipment to be used, preparing a lesson plan, or ‘managing’ learning activities for the students during the lesson. Seen as a sociocultural or discursive practice, ‘pedagogy’ is encompassing of knowledges, mind, ways of knowing, language and discourses; epistemologies of the learner and the
114
More pedagogies
teacher; educational goals, purposes, values, expectations, curriculum; the interactions and relationships between participants; the prior knowledge, motivation; the affective; the diversity of students as well as the more widely known facets of teachers, teaching, learners, learning, assessment (Leach and Moon, 1999b). Such a view of pedagogy also constructs the learner and the learner’s mind as being agentive, complex and multifaceted; the curriculum, not as a single entity, but as the planned, enacted and experienced curriculum; knowledge as identity as it is linked with students’ emotions, motivation and self-esteem; and it constructs the conditions and opportunities for reflection, productive cognitive conflict and the development of habits of mind (Leach and Moon, 1999a). The term ‘pedagogy’ is often preceded by an adjective, such as in the terms: • • • • • •
feminist pedagogies (Middleton, 1993); black/urban pedagogies (Delpit, 1993); critical pedagogies (Luke and Gore, 1992); Maori pedagogies (Bishop and Glynn, 1999; Hohepa, McNaughton and Jenkins, 1996; Metge, 1984; Pere, 1982; Smith, 1986); Pacific pedagogies (Manu’atu, 2000; Thaman, 1995, 1997); queer pedagogies (Pinar, 1998).
These adjectives (for example, radical, feminist, urban, critical, Maori, Pacific, queer) signal the educational goals, values, knowledges, ways of knowing etc. that are embodied in that specific use of the term ‘pedagogy’, in opposition to and in critique of those of the ‘mainstream’ or ‘traditional’ pedagogy (Gore, 1993). This use of the term ‘pedagogy’ is usually (but not always) associated with post-structuralist theorizing as in the Waikato doctoral theses (Gilbert, 1997; McKinley, 2003). There is also a second use of the term ‘pedagogy’, that is, teaching being viewed as a situated sociocultural practice, with origins in the psychology of learning, teaching, and assessment (Bell, 2003). The mainstream view of teaching had historically been one of process–product (Dunkin and Biddle, 1974), based on the psychological theorising of behaviourist views of learning, information processing views of learning, and cognitive science, all being based on individualism. However, recent research on teaching has tended to use sociocultural, discursive and poststructural theorising of learning, teaching and assessment (Bell, 2000). In using this theorising, there is acknowledgement of the social, political, and cultural dimensions of teaching, learning and assessment, and hence the more appropriate term ‘pedagogy’ is used. In addition, the use of the plural ‘pedagogies’ signifies the multiple theorising, practices, epistemologies, and ontologies that are constructed under the general umbrella called ‘pedagogy’. Both views of pedagogy, based on theorising from the sociology, philosophy and psychology of education, suggest that theorising of educational practice by practitioners is both developed and redeveloping. Theorising of pedagogy is also done by teachers in early childhood centres and schools, and by teacher educators,
More pedagogies
115
as we reflect and theorise our own teaching practice as teachers, and that of our students, pre-service teachers and in-service teachers. The debate is essentially focused on how self-study by teachers (primary, secondary and tertiary) can help us to develop our teaching practice, not just by reflection, but by theorising as well (Loughran, Hamilton, LaBoskey and Russell, in press; Loughran and Russell, 2002). The questions ‘whose knowledge or theorising counts?’; ‘for what purposes is the theorising done?’ are relevant here. These questions have been asked in other disciplines, for example, science (Harding, 1991) and are being asked in education too. It is hoped that the theorising of pedagogy in education academic texts, government documents and teacher education and teachers’ communities of practice will foster the development of theorising of ‘pedagogy’ for practitioners of education, as well as for the discipline of education, and not just those in the associated disciplines of sociology, philosophy and psychology. The first twenty years of science education research at the University of Waikato make a valuable contribution towards this.
116 Classroom Chapter 6 assessment of science learning
Classroom assessment of science learning
Introduction Assessment of classroom learning of science was increasingly addressed in the Waikato research in the 1980s and 1990s, reflecting a general trend in both education and science education (Bell, in press), and in part the political nature of assessment in education, including science education. For example, during the 1980s and 1990s, as in many other countries, assessment in education in New Zealand underwent substantial policy review and changes (Crooks, 2002b; Phillips, 1998), along with parallel national curriculum developments, including the national science curriculum (Bell, Jones and Carr, 1995). The main assessment developments have been: • • •
• •
an increase in the requirements for teachers to obtain assessment information for accountability purposes (Hill, 1999, 2000); the establishment of a national monitoring programme (Crooks and Flockton, 1996; Flockton, 1999); a restructuring of the national qualifications for school learners (Lee and Lee, 2001; Strachan, 2002) into the National Certificate of Educational Achievement (NCEA), including a move from norm-referenced to criterionreferenced system for assessment, with unit standards (leading to work-based qualifications) and achievement standards (based on the school curriculum subjects), and with internal and external assessment; an increased focus on using formative assessment, and feedback in particular, to improve learning outcomes (Bell and Cowie, 2001b; Hattie, 1999); increased resources for assessment by teachers (for example Chamberlain, 2001; Clarke, Timperley and Hattie, 2003; Gilmore and Hattie, 2001; Hattie, 2003).
It is within these political contexts that the research into assessment in science education was done at the University of Waikato. These developments resulted in teachers of science increasingly being asked to do assessment in their
Classroom assessment of science learning
117
classrooms for multiple purposes. And the number of these purposes has increased with the number of different shareholders wishing to use the assessment information generated by teachers and students in science classrooms for different purposes. These purposes are linked not only to the goals of education and science education but also to the political nature of assessment. This international trend, of multiple purposes for science classroom assessment, was brought into sharp focus in the 1990s, when politicians, and others wanting to hold educationists accountable, looked to assessment to provide the information required for the accountability process, for example to audit science teacher effectiveness (Bell and Cowie, 2001b). This added to the existing demands for assessment information by people who operate outside the classroom, for example caregivers, principals, school governing bodies, local or national government officials, awarders of national qualifications, selection panels for tertiary education programmes and employers. These multiple purposes can include auditing of schools, national monitoring, school leaver documentation, awarding of national qualifications, appraisal of teachers, curriculum evaluation and the improvement of teaching and learning. While these purposes are often mandated by people who operate outside the classroom, the assessments themselves are often done by teachers on their behalf. There are three cornerstones of this education accountability process – a prescribed set of standards, an auditing and monitoring process to ascertain whether the standards have been attained, and a way of raising standards if low standards have been indicated in the audits (Bell and Cowie, 2001b). Classroom assessment of (science) learning is seen as a way of raising standards (Atkin, Black, Coffey and National Research Council (US) Committee on Classroom Assessment and the National Science Education Standards, 2001). An indication of this trend of multiple purposes for assessment in New Zealand was evident by the early 1990s, when policy documents (Ministry of Education, 1990e, 1994a) advocated these three cornerstones of accountability. In New Zealand, the ‘standards’ are contained in the New Zealand Curriculum Framework and associated documents (Ministry of Education, 1993a, 1993b), including those of the New Zealand Qualifications Authority; the auditing is done by the Education Review Office (http://www.ero.govt.nz); and the monitoring by the National Education Monitoring Project (http://nemp.otago.ac.nz). The ‘raising of standards’ was seen by policy makers in New Zealand as being achievable by a number of methods, including school-based assessment (Ministry of Education, 1993a). This more recent addition of assessment for accountability purposes is reflected in this recent statement of the three main purposes for assessment in education: Assessment has multiple purposes. One purpose is to monitor educational progress or improvement. Educators, policymakers, parents and the public
118
Classroom assessment of science learning
want to know how much students are learning compared to the standards of performance or to their peers. This purpose, often called summative assessment, is becoming more significant as states and school districts invest more resources in educational reform. A second purpose is to provide teachers and students with feedback. The teachers can use the feedback to revise their classroom practices, and the students can use the feedback to monitor their own learning. This purpose, often called formative assessment, is also receiving greater attention with the spread of new teaching methods. A third purpose of assessment is to drive changes in practice and policy by holding people accountable for achieving the desired reforms. This purpose, called accountability assessment, is very much in the forefront as states and school districts design systems that attach strong incentives and sanctions to performance on state and local assessments. (National Research Council, 1999) The research at Waikato group has had a focus on assessment of classroom science learning, for formative (including diagnostic) and summative purposes, and is reviewed under the following headings: • • • • •
The assessment of multiple goals of science learning; Assessment for formative purposes; Assessment for summative purposes; The notion of quality in educational assessment; and Theorising of assessment.
The assessment of multiple goals of science learning As with the international trends, the research at Waikato has addressed the assessment of multiple goals of science education. ‘What is assessed?’ is an important question as it links assessment strategies, learning goals and curriculum. A strong criticism of assessment in the past has been that only learning goals that could be readily assessed, say by using recall to answer multiple-choice tests, were assessed, with a subsequent negative impact on the curriculum, pedagogy, learning and learners in the classroom (Crooks, 1988). There is now the recognition that all learning goals need to be assessed and not just recall and understanding of science concepts because they are easy to test for (Osborne and Ratcliffe, 2002). In the early 1980s, the assessment of conceptual understanding, process skills and attitudes was seen as important in New Zealand curriculum documents (Department of Education, 1978, 1979). However, the publications of the first Learning in Science Project (F1–4) added thinking or cognitive skills (a combination of the conceptual understanding and scientific skills) and social skills as science learning to be assessed, for example:
Classroom assessment of science learning
119
A teacher’s assessment list could include: (i) (ii)
(iii) (iv) (v)
Interpretive skills: Ability to plan, carry out and report on an investigation. Cognitive skills: Ability to formulate a sensible conclusion in terms of the evidence, to interpret results and make predictions, to appreciate the views of others, and the evidence on which it is based. Manipulative skills: Ability to set up gear, make readings, follow instructions where necessary. Workshop skills: Ability to tidy up after experiments, observe safety procedures. Social skills: Responsibility to group, class. Ability to share and listen to the views of others. ( Tasker and Freyberg, 1985: 79)
These skills were also being considered in the international literature (Millar and Driver, 1987). The assessment of thinking skills was also addressed in the initial publication of the Learning in Science Project (Primary): . . . Thinking skills – In the (first) LISP work our attempt to do research on specific process skills was unsuccessful for reasons already mentioned. I would propose that the non-psychomotor process skill we should investigate is the skill of investigating or problem solving . . . (Osborne, 1982b: 5) The project went on to research children’s classroom investigations, based on the questions asked by the students (Osborne and Biddulph, 1985b) and thinking skills and social skills were included in later curriculum documents (Ministry of Education, 1990a, 1993a,b). By 1993, an international review of assessment of science learning listed what was worth assessing as knowledge of facts and concepts, science process skills, higher order science thinking skills, problem-solving skills, skills needed to manipulate laboratory equipment and attitudes of science (Doran, Lawrenz and Helgeson, 1993). Likewise, since the early Learning in Science Projects, the additional curriculum science learning goals needing to be assessed, as reported in the Waikato research, include: • • • • •
aspects of practical work (Lal, 1991; Saeed, 1997; Torrie, 1989); procedural knowledge (English and Wood, 1997); investigations and problem solving (Haigh, 1998; Haigh and Hubbard, 1997); the nature of science views held by students (Hipkins and Barker, 2002; Loveless and Barker, 2000); knowing that science is culturally and historically embedded, and contextualised (Barker, 1997);
120
•
•
Classroom assessment of science learning
learning dispositions (habits of mind), such as resilience, playfulness, reciprocity, curiosity, friendliness, being bossy, confidence, curiosity, intentionality, self-control, relatedness, communication, cooperation, courage, playfulness, perseverance, responsibility, selectivity, experimentation, reflection, opportunism, conviviality, (Carr, 2001; Carr and Claxton, 2002) as well as effective learning skills in the learning-to-learn literature (Baird and Northfield, 1992); personal and social development (Cowie, Boulter and Bell, 1996).
For example, the 10 teachers in the Learning in Science Project (Assessment) commented that they did more than assess the scientific content of the students’ learning. They indicated that the students’ learning within the classroom involved their personal and social development/learning as well as their science development. Students’ personal development related to their learning about themselves as learners of science, asking questions, self-assessment, behaviour, time management, motivation and attitude. Their social development related to their interacting with others (students and teachers), peer assessment, leadership skills, group work, discussion and listening skills. Their science development related to the development in the knowledge and understanding of science and their ability to do science – for this was their unique purpose for being in a science classroom. These three aspects were assessed by the teachers of science (Cowie and Bell, 1996a). The three aspects were not independent of each other; the complexity and richness of their interactions was a contributor to the diversity and complexity of the classroom. The three aspects are conceptualised as three intersecting circles on the left-hand side of Figure 6.1: what is assessed? The aspects of science which were assessed in the science lesson are represented on the right hand side of Figure 6.1, namely science content (the body of
social development
personal development
science content
science learning
Figure 6.1 What is assessed in science classrooms?
science context
science processes
Classroom assessment of science learning
121
scientific knowledge – the concepts and ideas of science), science context (the contexts in which the science is learnt and used) and science processes (those skills and processes used by scientists to investigate phenomena). The balance of assessment of personal, social and science development or learning was discussed by the teachers on the teacher development days. It was acknowledged that the three aspects may differ in weighting within a lesson and within a unit of work; with different learning goals for a lesson and with different abilities and ages of the students. The analogy was made between the circles and balloons that inflate and deflate. The teachers’ comments indicated that the way they assessed the social, personal and science aspects was different. The secondary teachers felt that science assessment was done more by formal methods and the assessment of the social and personal more by informal. The primary teachers indicated that they assessed, by formal ways, both the personal and social learning as well as the science learning.
Assessment for formative purposes Definition and characterisics Formative assessment is increasingly being used to refer only to assessment which provides feedback to students (and teachers) about the learning which is occurring, during the teaching and learning, and not after (Cowie, 1997). The feedback or dialogue is seen as an essential component of formative assessment interaction where the intention is to support learning (Black and Wiliam, 1998a; Gipps, 1994a; Hattie and Jaeger, 1998). And assessment can be considered formative only if it results in action by the teacher and students to enhance student learning (Black, 1993). These components are reflected in various definitions of formative assessment, for example, ‘The process used by teachers and students to recognise and respond to student learning in order to enhance that learning, during the learning’ (Bell and Cowie, 2001b: 8). It is through the teacher–student interactions during science and other learning activities (Newmann, Griffin and Cole, 1989) that formative assessment is done, that students receive feedback on what they know, understand and can do, and receive teaching to learn further. Formative assessment is at the intersection of teaching and learning (Gipps, 1994a) and in this way teaching, learning and assessment are integrated in the curriculum, including the science curriculum (Hattie and Jaeger, 1998). The term ‘formative interaction’ ( Jones et al., 2003; Moreland et al., 2001) may be used instead of ‘formative assessment’ to highlight this interactive nature of formative assessment – that teacher–student interactions are the core of formative assessments. Science assessment for diagnostic purposes, for example (Barker and Carr, 1989b), is therefore included as is embedded assessment of science learning (Treagust, Jacobowitz, Gallagher and Parker, 2001). Harlen and James (1997), in a review of the general assessment
122
Classroom assessment of science learning
literature, summarised that the characteristics of formative assessment, to distinguish it from summative assessment, are: • • • • • •
it is positive; a part of teaching; takes into account the progress of students; can elicit inconsistencies which can provide diagnostic information; places more value on validity and usefulness than reliability; requires the students to be actively involved in monitoring their own progress and improving their learning. Harlen (1998) also described assessment for formative purposes as that which [is] embedded in a pedagogy of which it is an essential part; shares learning goals with students; involves students in self-assessment; provides feedback which leads to students recognising ‘the gap’ and closing it; [is] underpinned by confidence that every student can improve; and involves reviewing and reflecting on assessment data. (Harlen, 1998: 3)
Bell and Cowie (1997b, 2001b), in reporting the findings of the Learning in Science Project (Assessment), summarised the nine characteristics of formative assessment of science learning, as discussed by the teachers, as follows: •
• • • • • • • •
it is responsive (that is, dynamic and progressive, informal, interactive, unplanned as well as planned, proactive as well as reactive, responding with individuals and with the whole class, involves uncertainty and risk taking, and has degrees of responsiveness); it uses written, oral and non-verbal sources of evidence; it is a tacit process; it involves student disclosure; it uses professional knowledge and experiences; it is an integral part of teaching and learning; it is done by both teachers and students to improve teaching as well as learning; it is highly contextualised; and it involves managing dilemmas.
Importance of formative assessment Formative assessment, like assessment in general, does influence learning (Crooks, 2002a; Gipps and James, 1998). The case for formative assessment was made in a report commissioned by the British Educational Research Association to argue the case for raising achievement through the use of assessment for formative purposes, rather than through large-scale testing for accountability purposes. This seminal review by Black and Wiliam (1998a) of the research documented in
Classroom assessment of science learning
123
578 articles, stated the importance of formative assessment for learning as ‘The research reported here shows conclusively that formative assessment does improve learning. The gains in achievement appear to be quite considerable, and as noted earlier, amongst the largest ever reported for educational interventions’ (p. 61). Likewise, Hattie (1999) concluded his meta-analysis to evaluate the relative effects of different teaching approaches and different components of teaching by stating that the most powerful single moderator that enhances achievement is feedback. Having reviewed the literature to document the evidence that formative assessment can indeed raise standards Black and Wiliam (1998a) then addressed the question of ‘is there evidence that there is room for improvement?’ They concluded that there is research evidence that ‘formative assessment is not well understood by teachers and is weak in practice; that the context of national or local requirements for certification and accountability will exert a powerful influence on its practice; and that its implementation calls for rather deep changes both in teachers’ perceptions of their own role in relation to their students and in their classroom practice’ (Black and Wiliam, 1998a: 20). The science education literature to support this knowledge claim is found in Bol and Strage, 1996; Duschl and Gitomer, 1997; Lorsbach, Tobin, Briscoe and LaMaster, 1992. Despite there being much advocacy by science educators on the importance of formative assessment to improve learning and standards of achievement, there have been only a few research studies on the process of formative assessment of science learning. A model of formative assessment In the research done as part of the Learning in Science Project (Assessment), ten teachers (five intermediate and five secondary teachers) generated data about the assessment they felt they were doing for formative purposes. They spoke of doing two kinds of formative assessment, which they termed, ‘planned’ and ‘interactive’ formative assessment (Bell and Cowie, 1997b, 2001b). The planned formative assessment was teacher directed – the teacher having agency and the teachers’ purposes related to teaching being addressed. The main purpose for which the teachers said they used planned formative assessment was to obtain information from the whole class about progress in learning the science as specified in the curriculum, that is, the feedback was for the teacher’s purposes. This form of formative assessment was planned by the teacher mainly to obtain feedback to inform her or his teaching. This assessment was planned in that the teacher had planned to undertake a specific activity (for example, a survey or brainstorming) to obtain assessment information on which some action would be taken. The teachers considered the information collected as a part of the planned formative assessment was ‘general’, ‘blunt’ and concerned their ‘big’ purposes. It gave them information which was valuable in informing their interactions with the class as a whole with respect to ‘getting through the curriculum’.
124
Classroom assessment of science learning
The second kind was interactive formative assessment. This kind had more student agency as the teachers were responding to something said or done by the students, and students’ purposes related to learning being addressed. The feedback given is mainly for the students to gain awareness as to their learning, although the feedback was also of use to the teachers regarding their teaching. Interactive formative assessment was that which took place during student–teacher interactions. It differed from the first form – planned formative assessment – in that a specific assessment activity was not planned. The interactive assessment arose out of a learning activity. Hence, the details of this kind of formative assessment were not planned, and could not be anticipated. Although the teachers often planned or prepared to do interactive formative assessment, they could not plan for or predict what exactly they and the students would be doing, or when it would occur. As interactive formative assessment occurred during student–teacher interaction, it had the potential to occur any time students and teachers interacted. The teachers and students within the project interacted in whole-class, small-group and one-to-one situations. It was during the interactive formative assessment that students gained feedback on their learning and feedforward to promote their learning. The main purpose for which the teachers said they did interactive formative assessment was to mediate in the learning of individual students with respect to science, social and personal learning. Hence, they said they formatively assessed a wider range of learning outcomes than the science. The teachers’ specific purposes for interactive formative assessment emerged in response to what sense they found the students were making. The teachers indicated that through their interactive formative assessment, they refined their short-term goals for the students’ learning within the framework of their long-term goals. For example, Teacher 5 changed her purposes for learning, within a unit of work on earth sciences, from learning about weathering to learning about contracting and expanding when she noticed some of the students had scientifically unacceptable conceptions about heating and cooling. The teachers indicated that their purposes for learning could be delayed. For example, Teacher 7 delayed the learning about separating mixtures until the students had learnt about the properties of the substances to be separated. The purpose of student learning was negotiated between the teacher and the students through formative assessment feedback. Teacher 9 described it as linking students into her agenda. The teachers described interactive formative assessment as teacher- and student-driven rather than curriculum-driven. They said the focus of their interactive formative assessment was ‘finer tuned’ with ‘lots of little purposes to support the major picture or purpose’. Interactive formative assessment was therefore embedded in and strongly linked to learning and teaching activities and links to later research and development on feedback and goal setting for students (Clarke, 1998, 2001, 2003; Clarke et al., 2003).
Classroom assessment of science learning
125
eliciting
acting
purpose
interpreting
Figure 6.2 Planned formative assessment noticing
responding
purpose
recognising
Figure 6.3 Interactive formative assessment
The process of planned formative assessment was characterised by the teachers as eliciting, interpreting and acting on assessment information. These aspects can be represented diagrammatically as in Figure 6.2. Each of the four parts of the planned formative assessment process are described in more detail in Bell and Cowie, 1997b, 2001b. The process of interactive formative assessment involved the teachers noticing, recognising and responding to student thinking during these interactions and can be represented diagrammatically as in Figure 6.3. Each of the four parts of the interactive formative process are described in more detail in Bell and Cowie, 1997b, 2001b. The two kinds of formative assessment and the links between them, as seen by the teachers, can be represented diagrammatically as in Figure 6.4. The teachers discussed the relationship between and the interaction of the two kinds of formative assessment (for a fuller account, see Bell and Cowie, 1997b: 314). They commented that the two kinds of formative assessment were linked through the purposes of formative assessment (see the dotted line); that some
126
Classroom assessment of science learning
eliciting
purpose
acting
noticing
interpreting
purpose
recognising
responding
planned formative assessment
interactive formative assessment
Figure 6.4 A model of formative assessment
teachers used interactive formative assessment more than other teachers; and that a teacher moved from planned to interactive and back. The link between the two parts of the model was seen by the teachers to be centred around the purposes for doing formative assessment. The teachers confirmed that they changed from planned to interactive formative assessment by noticing something in the course of planned formative assessment. They may have suspected that things may not have been okay and wanted to check things out; they may have noticed a student’s or a group of students’ alternative conceptions or misconceptions; they may have wished to follow up a hunch, or monitor the learning occurring. This change was usually in response to focusing from the class to an individual. They usually switched back from interactive to planned formative assessment in response to their responsibility for the learning of the whole class. They also commented that under stress (for example, implementing a new curriculum or when ill) they tended to do less of the interactive formative assessment. In particular, heavy emphases on summative assessment procedures for leaving qualifications (for example, Unit Standards in New Zealand) or for review and monitoring procedures (by the Education Review Office in New Zealand) were seen by the teachers as influencing the amount of interactive formative assessment they felt they were able to do. The main difference and similarities between the two forms of formative assessment are given in Table 6.1. The key features of the model are now discussed, in association with other research on formative assessment. This model is notable in that it was developed by the teachers involved in the research project as part of the research process (Bell and Cowie, 1999). They were asked to develop a model that would communicate to teachers not involved in the research what it was that they were doing when they were doing formative assessment. The primary teachers
Classroom assessment of science learning
127
Table 6.1 Planned and interactive formative assessment Planned formative assessment
Interactive formative assessment
The parts of the process were eliciting, interpreting, acting
The parts of the process were noticing, recognising, responding
Tended to be done with all the students in the class
Tended to be done with some individual students or small groups
Could occur over an extended time frame
Happened over a short time frame
Purposes were mainly science-referenced
Purposes were science-, student- and care-referenced
Responsive to ‘getting through the curriculum’
Responsive to student learning
What was assessed was mainly science learning
What was assessed was science, personal and social learning
The assessment information obtained was product and process
The assessment information obtained was product and process but ephemeral
Interpretations were norm-, science- and student-referenced
Recognising was science-, norm- and student-referenced
Actions were science-, student- and carereferenced
Responses were science-, student- and care-referenced
Relied on teachers’ professional knowledge
Relied on teachers’ professional knowledge
in the research by Torrance and Pryor (2001) similarly developed a ‘practical classroom model’. A second key feature of the model is the central role given to purpose in both forms of formative assessment by the teachers. It is also of interest to note that the teachers in the Torrance and Pryor (2001) research placed the ‘making task and quality criteria explicit’ in the centre of their classroom practice model, which is linked to purpose. Thirdly, the model of formative assessment developed by the science teachers included the notion of planning which has also been highlighted by other science education researchers (Harlen, 1995) and others ( Torrance and Pryor, 1995). A fourth key feature of the model is that formative assessment is described as a complex, highly skilled task, as it is in other research ( Torrance and Pryor, 1998), and a task which relied on the following knowledge bases (Shulman, 1987) of content knowledge, for example knowing the scientific understanding of the concepts being taught; general pedagogical knowledge, for example of classroom management; curriculum knowledge, for example of the learning objectives in the curriculum being taught; pedagogical content knowledge, for example knowing how best to teach atomic theory to a class of 14-year-olds; a knowledge about learners in general and the students in the class; knowledge of educational contexts, for example the assessment practices in the school; a knowledge of educational
128
Classroom assessment of science learning
aims and purposes, for example a possible ‘science-for-all’ emphasis in a national curriculum. To this list, the teachers’ knowledge of progression in students’ learning of specific science concepts can be added (Bell and Cowie, 2001c; Jones et al., 2003). The formative assessment also relied upon the processes of pedagogical reasoning and action (Shulman, 1987), including the transformation of content knowledge into pedagogical knowledge, through preparation, representation, selection and adaptation. The teachers felt the use of both forms of formative assessment and the switching between them was the hallmark of a competent teacher. A fifth key feature of the model is teachers’ interaction skills and the nature of the relationships they had established with the students was also seen as important. It was felt that the teachers needed a disposition to carry out interactive formative assessment, that is, the teachers needed to value and want to interact with the students to find out what they were thinking about. Cowie (2000) also commented on the relationships that developed between the teachers and students as individuals, groups and as a class. Mutual trust and respect were identified as the key factors mediating student willingness to disclose their ideas to teachers and peers, and hence enable formative assessment interactions to occur. Hence the findings support the contention by Tittle (1994) that the views and beliefs of the interpreters and users of assessment information (here teachers and students) are an important dimension of any theory of educational assessment. A sixth key feature of the model is the action taken as part of both planned and interactive formative assessment, for it distinguishes assessment for formative purposes from that for summative and accountability. The action means that formative assessment can be described as an integral part of teaching and learning and that it is responsive to students. Much of the current literature, for example Driver, Squires et al., 1994, on conceptual development in science education involves a consideration of the teacher being responsive to the thinking of students, often phrased as ‘taking into account students’ thinking’. To respond to and mediate students’ thinking involves the teacher finding out what the thinking is, evaluating the thinking and responding to it. These are the three components in both planned and interactive formative assessment. The teachers in the research made the claim that they did not think they could promote learning in science unless they were doing formative assessment (Bell and Cowie, 1997b). The role of the teacher included providing opportunities for formative assessment to be done (for example, having the students discuss in small groups their and scientists’ meanings of ‘electric current’ rather than listening only to a lecture by the teacher), and using the opportunity to do formative assessment (for example, interacting with the students whilst they are doing small-group discussion work about their conceptual understandings of electric currents). In addition, the action taken as part of both planned and interactive formative assessment was seen by the teachers as a part of teaching and by the students as a part of learning. The teachers acted and responded on the assessment information they obtained in science (criterion-referenced), student (ipsative) and care-referenced ways. In the care-referenced actions, the teachers took action to sustain and enhance the quality of interactions and relationships between the students and between them-
Classroom assessment of science learning
129
selves and the students. Other research has also noted the dual ipsative and criterion-referenced nature of formative assessment (Harlen and James, 1997) and the care aspect of formative assessment (Treagust et al., 2001). A single action or response might have one or more of these aspects in it. It was the action part of planned and interactive formative assessment that the teachers felt they needed more help with in future teacher development. One important aspect of the ‘taking action’ is the feedback to the student from the teacher or another student. The feedback is more effective in improving learning outcomes if it is about the substance of the work, not superficial aspects (Harlen, 1999); linked with goal setting (Gipps and Tunstall, 1996b; Hattie and Jaeger, 1998); linked to the student’s strengths and weaknesses in doing the task, rather than just to the self, as in praise (Hattie, Biggs and Purdie, 1996). The quality of the feedback may involve a comparison between the student’s achievement or performance and other students (norm-referenced); standards or learning goals (criterion-referenced), or the student’s previous achievements (ipsative). In assessment for formative purposes the ipsative frame of references for feedback is important. Another important aspect of the taking action part of the formative assessment process has become known as ‘feedforward’, to distinguish it from feedback (Bell and Cowie, 2001b). Whilst feedback was used to refer to the response given to students by the teacher (Sadler, 1989) or another student about the correctness of their learning, the phrase ‘feedforward’ is used to refer to those aspects of formative assessment in which the teacher was helping the students to close the gap between what they know and can do, and what is required of them as indicated in the standards or curriculum objectives (Sadler, 1989). Hence, to provide both feedback and feedforward a teacher must know the curriculum content and standards or curriculum objectives, the progression of students’ learning, and the scaffolding required for learning in the Zone of Proximal development, after Vygotsky (Torrance and Pryor, 1998). A seventh key feature of the teachers’ model of formative assessment was the central role of self-assessment and self-monitoring. This is distinct from feedback which is given by another person (Sadler, 1989). Research on this aspect of assessment for formative purposes was also reviewed in the meta-analysis by Hattie et al. (1996), who concluded that interventions, which are integrated to suit the individual’s self-assessment, orchestrated to the demands of the particular task and context, and self-regulated with discretion, were ‘highly effective in all domains (performance, study skills and affect) over all ages and abilities, but were particularly useful with high-ability and older students’ (p. 128). And they were more effective than the typical study skills training packages. To be able to give effective feedback and feedforward, the research with teachers of technology indicated that the pedagogical content knowledge as well as pedagogical approaches of teachers had to be enhanced (Moreland and Jones, 2000). This is so that the teachers could make a judgement about where the student’s learning is in relation to the intended curriculum learning goals, communicate this to the student, and suggest steps for the student to improve his or her learning, based on his/her knowledge of progression in learning a specific skill or concept.
130
Classroom assessment of science learning
The following have been suggested for teachers and students, on the basis of reviews of the general assessment literature research evidence, as interventions to improve the use of assessment for formative purposes (Black and Wiliam, 1998b: 9–13): Feedback to any pupil should be about the particular quality of his or her work, with advice on what he or she can do to improve, and should avoid comparisons with other pupils. For formative assessment to be productive, pupils should be trained in selfassessment so that they can understand the main purposes of their learning and thereby grasp what they need to achieve. Opportunities for pupils to express their understanding should be designed into any piece of teaching, for this will initiate the interaction whereby formative assessment aids learning. The dialogue between pupils and a teacher should be thoughtful, reflective, focused to evoke and explore understandings, and conducted so that all pupils have an opportunity to think and to express their ideas. Tests and homework exercises can be an invaluable guide to learning, but the exercises must be clear and relevant to learning aims. The feedback on them should give each pupil guidance on how to improve, and each must be given opportunity and help to work at the improvement. This was later summarized to: The research indicates that improving learning through assessment depends on five, deceptively simple key factors: the provision of effective feedback to pupils; the active involvement of students in their own learning; adjusting teaching to take account of the results of assessment; a recognition of the profound influence assessment has on the motivation and self-esteem of pupils, both of which are crucial influences on learning; and the need for pupils to be able to assess themselves and understand how to improve. (Assessment Reform Group, 1999: 4) and: Sharing learning goals with pupils; involving pupils in self-assessment; providing feedback which leads to pupils recognizing their next steps and how to take them; underpinned by confidence that every pupil can improve. (Assessment Reform Group, 1999: 7) Publications have been produced for teachers, based on the above research reviews, to help them improve the assessment for formative purposes in the classroom (including the science classroom) (Atkin et al., 2001; Clarke et al., 2003). For example, the last publication has a chapter on each of the components of formative assessment as:
Classroom assessment of science learning
131
clarifying learning intentions at the planning stage, as a condition for formative assessment to take place in the classroom; sharing learning intentions at the beginning of lessons; involving children in self-evaluation against learning intentions; focusing oral and written feedback around the learning intentions of lessons and tasks; organizing individual target setting so that children’s achievement is based on previous achievement as well as aiming for the next level up; appropriate questioning; and raising children’s self-esteem via the language of the classroom and the ways achievement is celebrated. (Clarke et al., 2003: 14) The last key feature of the teachers’ model of formative assessment is the teacher development that occurred during the development of the model by the teachers (Bell and Cowie, 1997b, 2001c), providing some information to answer Black and Wiliam’s question ‘Is there evidence about how to improve formative assessment?’ Information has also been provided by other researchers, using collaborative action research ( Torrance and Pryor, 2001), reflective surveys (Black and Harrison, 2001a, 2001b) and reflection on teachers’ knowledge bases ( Jones et al., 2003). A notable feature of the literature is that teacher development for assessment for formative purposes also involves changing one’s overall pedagogy, not just the assessment aspects (Ash and Levitt, 2003; Bell and Cowie, 2001c; Black and Wiliam, 1998a). Students’ views of formative assessment There is a growing interest in the wider education research literature in the views of students on teaching, learning and assessment (as distinguished from their views on the subject matter content), for example Morgan and Morris (1999). Whilst there have been reviews on the impact of assessment on students (Crooks, 1988; Hattie and Jaeger, 1998), there has been little research until recently on students’ views of assessment, for example Brookhart (2001); Gipps and Tunstall (1996a); Pollard, Triggs, Broadfoot, McNess and Osborn (2000). A Waikato research study is worthy of discussion in this context (Cowie, 2000). In this doctoral research, 75 students (Years 7 to 10, or ages 11 to 14) were interviewed in either individual or group situations about their views on classroom assessment. The findings indicate that the students constructed themselves as active and intentional participants in learning, its assessment and their selfassessment of it. The criteria for judging the success of their learning, reported by students, included the ability to perform a task, gaining good marks or grades, the teacher confirming their ideas were correct and feelings of completeness and coherence. Another finding was that students viewed formative assessment as embedded in and accomplished through interaction with teachers, peers and parents. Disclosure was another aspect of students’ views of formative assessment. The students were very aware that their questions, actions and bookwork
132
Classroom assessment of science learning
had the potential to disclose not only what they knew but also what they did not know, to peers and teachers, who may or may not make positive judgements and take action on the basis of these disclosures. Students indicated that they withheld disclosure if the classroom was not safe, only disclosing if they were in a trusting relationship with the teacher and peers. This would influence not just the validity but the essence of formative assessment. Student disclosure is central to formative assessment and participation in assessment interactions could lead to both benefit and harm in learning, social and relationship constructions. Torrance and Pryor (1998) described the teacher during formative assessment as using power-with and power-for students in their learning. Browen Cowie (2000) also detailed the ways in which student perspectives of formative assessment in the classroom contributed to the mutual construction of what it means to be a student and a teacher in that classroom, that is, notions of identity. For example, ‘student perceptions that time and attention were limited and teachers assessed what was important to them, meant that teacher assessment served to communicate to the students who and what was important to the teacher’ (p. 260); for students, the key feature of formative assessment as a meaning making activity was that it contributed to their identity in the classroom . . . the students contended that disclosing their ideas in an attempt to enhance their understanding could lead their peers and the teacher to perceive them as ‘try hards’, ‘bright’ or ‘dumb’ and to their learning being enhanced or them being embarrassed and feeling stupid. They indicated that for them, assessment and learning were intimately connected and inherently linked with who they were and how they felt (p. 261). The students and teachers in Cowie’s study were seen as actors (that is, taking action) in formative assessment. The students were actors in formative assessment in three ways: their academic and social goals and interests mediated their interactions; they sought to manage the disclosure of their learning by choosing (or not choosing) to ask questions and by acting to restrict teachers’ incidental access to their bookwork; and they assessed the teacher to ascertain how the teacher reacted to their questions and therefore to find out what was seen as important by the teacher. In summary, Cowie (2000) stated that the students were both active in the formative assessment process and profoundly affected by it as did Reay and Wiliam (1999). The students in Cowie’s (2000) study insisted that their teachers could only assess their learning through face-to-face interaction with them. Face-to-face interaction was considered to enhance the fidelity (Wiliam, 1992) of teacher formative assessment, because students could negotiate the meaning of teacher questions and because students were more prepared to ask questions, thereby disclosing their views. Teachers
Classroom assessment of science learning
133
were said to provide more useful feedback during one-to-one and small group interactions. (Cowie, 2000: 267) Students’ views of assessment are also embedded in the use of questions generated by students as assessment information for diagnostic and formative purposes (Biddulph, 1989).
Assessment for summative purposes The main discussion on assessment for summative purposes in the 1980s and early 1990s Waikato research was the notion of assessment to indicate conceptual change or conceptual development. The debates were largely centred around the undertaking of pre- and post-teaching assessments to ascertain the change or growth in conceptual understanding, for example Bell (1984a), Bell and Barker (1982). The comparison of the pre- and post-assessments [also called ‘before-views’ and ‘after-views’ (Faire and Cosgrove, 1988) ] was a key aspect of assessment of conceptual development as was the choice of assessment tool to elicit the assessment data to be compared, for example: • • • • •
interviewing (Bell, 1995; Bell, Osborne, Tasker, 1985); interview-about-instances (Osborne and Gilbert, 1979, 1980a); interview-about-events (Osborne, 1980b; Osborne and Cosgrove, 1983); surveys based on children’s alternative conceptions (Freyberg and Osborne, 1985; Stead (now Bell) and Osborne, 1980); concept maps (Happs, 1984).
The assessment of change in conceptual understanding is different from the assessment of intended curriculum outcomes. As Driver and Easley (1978) described, there are two distinct approaches to research (and assessment) – the nomothetic and the ideographic. The nomothetic approach involves studies in which the pupils’ understanding is assessed in terms of the congruence of their responses with ‘accepted’ scientific ideas. In contrast, ideographic studies are those in which pupils’ conceptualisations are explored and analysed on their own terms without assessment against an externally defined system. Driver and Easley (1978) also distinguished between the two terms of ‘misconceptions’ and ‘alternative frameworks’. Misconceptions (with the obvious connotation of a wrong idea) is a term often used in studies where pupils have been exposed to formal models or theories, and have assimilated them incorrectly. Alternative frameworks arise when pupils develop autonomous concepts for conceptualising their experience of the physical world. This distinction is well recognised in the literature, for example Pfundt and Duit, 1994. Bell (1995a: 350) stated that if a teacher is taking into account students’ thinking and wishes ‘. . . to assess students’ learning in terms of conceptual change,
134
Classroom assessment of science learning
information on students’ thinking is needed in three situations: before, during and after teaching . . .’. Examples of questions to ask before the teaching are: • • •
What are the students’ ideas prior to teaching – their existing conceptions on the topic or ‘before-ideas’? What are the genuine questions students are likely to have about this topic? What are the students’ possible answers to some of their own questions? (p. 350)
During the teaching: • • • •
What questions are the students asking? Are they genuine and puzzling ones? What possible explanations or answers are the students proposing? Are the meanings that the students are constructing similar to the intended ones? Are the students changing their conceptions? (p. 351)
After the teaching: • • •
What are the students’ ideas when the teaching episode has finished (their ‘after-views’)? How do the students’ after-views compare with their before-views? That is, what conceptual change has occurred? How do the students’ after-views compare with the intended learning outcomes in the curriculum schemes? (pp. 351–2)
Research findings on assessment of conceptual development were also documented in the Learning in Science Project ( Teacher Development) (1990 –1992). In this research, the teachers discussed assessment with respect to its use as a criterion to indicate that they were ‘improving’ their teaching. When the teachers talked in interview about the classroom feedback they received on the new teaching activities from the students, the teachers often talked about the ‘better learning’ that was occurring (Bell and Pearson, 1992a). The new teaching activities were those that took into account students’ thinking: Teaching based on a constructivist view of learning, is defined here as teaching that takes into account students’ thinking and in particular involves . . . assessing the change and growth in students’ ideas, as well as the extent to which they had learnt the scientific ideas. (Bell and Gilbert, 1996: 10–11)
Classroom assessment of science learning
135
They stated that the ‘better learning’ was a factor that helped the teacher development. Two aspects of ‘better learning’ were discussed by the teachers: •
•
better conditions for learning, including increased enjoyment, social cooperation, ownership, student confidence and student motivation. If these conditions were present then the teachers felt learning was more likely to occur; better learning outcomes, including learning skills, conceptual development and improved results on tests and examinations. (Bell and Gilbert, 1996; Bell and Pearson, 1992a)
It appeared that the teachers in this study focused more on the learning conditions than the learning outcomes, as an indication of more effective teaching activities. By the late 1980s, internal assessment for summative purposes of practical work in the years 11–13 senior science subjects (biology, chemistry and physics) in New Zealand and internationally was being promoted as some of the learning goals associated with practical work were thought not to be validly assessed by the end-of-year external pen-and-pencil examinations. Hence, New Zealand, like other countries, saw a move from traditional written examination, to internal assessment of written reports of laboratory work, to achievement-based assessment. In 1989, Ian Torrie, then based at the University of Waikato, with five local teachers, developed a set of grade-related criteria for use in achievement-based assessment in Year 11 (Form 6) chemistry students in New Zealand ( Torrie, 1989). At the time, the assessment of practical work in Year 11 was for the national qualification of Form Six certificate, in which assessment overall was on a 1–9 point scale, with the award being moderated by the results of the norm-referenced examinations for these students from the previous year’s national examination. Torrie reported that parameters for the development of the grade-related criteria (GRC) articulated that the GRC needed to: 1 2 3 4 5 6
describe achievements in positive non-comparative language; discriminate between clearly different levels of achievement; provide useful information to students on what is expected and how they have fared; reflect the objectives of the course; assist in the assessment of all those aspects of the course that teachers need to assess; satisfy the goals of descriptive reporting of students’ achievements. (p. 287) The GRC developed were on a 1–5 grade scale (low=1, high=5) on the six criteria of knowledge, planning, performing, interpreting, communicating,
136
Classroom assessment of science learning
and teamwork. For example, the criterion of ‘performing’, was assigned these descriptors for the five grades: 1 2 3 4 5
attempts to use procedures; uses procedures; demonstrates some skill in using procedures; demonstrates skill in using procedures; demonstrates a high level of skill in using procedures. (p. 290)
In conclusion, Torrie noted that ‘the use of grade-related criteria has led to a greater focus on course objectives in teaching programs with skills rather than content being emphasised’ (p. 289), and ‘The need to match assessment criteria to syllabus statements has highlighted the need for the latter to offer very clear guidelines as to course objectives, levels of difficulty of content and the relative importance of the various course statements’ (p. 289). However, of key importance was that the teachers reported difficulty doing GRC assessment in the overall norm-referenced system of the Sixth Form Certificate. For example, Torrie noted that ‘even after a year’s involvement with the program, the teachers still found difficulty in accepting that the descriptor for Level 2 was not automatically half the value of the descriptor for Level 4’ (p. 289). Following this development work, one thesis (Lal, 1991) researched what teachers of Year 12 senior chemistry reported they were doing to assess practical work, based on the following research questions: 1 2 3 4 5
What are the current assessment methods used for assessing sixth form chemistry practical work? What are the teachers’ attitudes towards current assessment strategies in sixth form chemistry? What are the teachers’ views on assessment of an individual’s skill in group assessment? What are the teachers’ feelings on moderation of sixth form chemistry practical work? What are the teachers’ attitudes towards having a single grade for chemistry?
The self-report data was obtained by questionnaires sent to secondary schools in Auckland, Waikato, and Canterbury regions (n=158 returned from 212 sent by post). Follow-up interviews were conducted with 12 teachers from the Auckland and Hamilton regions. The findings included the following: 1
The common methods currently used for assessing practical work by teachers in the study are impression marking of laboratory performance and marking of laboratory reports.
Classroom assessment of science learning
2 3
4
5
137
The majority of the teachers were assessing practical work on the basis of written reports rather than by direct observation of pupils’ performance. The teachers in the study were in favour of profile reporting as a single grade for chemistry does not provide an adequate representation of pupils’ practical skills. The majority of the teachers in the study indicated that it was not possible to assess an individuals pupil’s practical ability from his or her performance in group tasks. The teachers in the study saw a need for some form of moderation between schools considering the range of assessment methods being used within schools.
In the 1990s, teachers were again expressing their concern about the implementation of a new policy on assessment for summative purposes for national qualifications. In this decade, much of the concern was due to the mix of criterion- and norm-referenced assessment for national qualifications with the introduction of unit standards and the achievement standards of the National Certificate in Educational Achievement (NCEA). However, there was also the concern about the use of students’ assessment information for staff appraisal purposes under the new political moves for increased accountability. The 10 teachers involved in the Learning in Science Project (Assessment) (Bell and Cowie, 2001b) expressed their concerns about assessing for summative purposes during baseline interviews about assessment of classroom learning in the initial phase of the research (Cowie and Bell, 1995). These teachers commented on: • • • • •
scientific knowledge being only one of the learning outcomes they valued within the classroom; coping with the number of individuals to be assessed and the complexity of the information to be collected and exchanged; the place and purpose of formal assessment (is it just for accountability or for improving children’s learning?); the tension between individual assessment and group work; the problem of representative reporting.
The 30 students who were interviewed as part of the Learning in Science Project (Assessment) raised a set of issues which were different but complementary to those of the teacher, namely: • • • • • •
the requirement on students to demonstrate their learning to others; maximising the use of written comments in books; students’ need to understand the teaching agendas of teachers; the usefulness of tests for review, but their perceived lack of fairness and representativeness; the place and use of self-assessment; students’ needs for feedback they can act on.
138
Classroom assessment of science learning
The teachers also expressed their concerns about assessment for summative purposes during the discussions on the teacher development days (Bell and Cowie, 1997b, 2001c). In 1995, the first year of the project, the teachers’ main concern was how formative assessment interfaced with the other assessment developments and issues in schools at the time. At their request, speakers from the Ministry of Education and the Education Review Office were invited to clarify for them policies and developments in assessment which affected what they as teachers did in the classroom. The main concerns about assessment in general were recording and reporting, especially to parents; assessments used for the purposes of the Education Review Office’s effectiveness reviews; some Ministry of Education’s policies on assessment; and assessment of learning with reference to the science curriculum. Recording and reporting was of concern to teachers, including: the number of assessments to be recorded for reporting purposes; how reports can become more representative of what students can actually do; the reporting of progress; comparing students with each other in reports and the validity and reliability of recording and reporting. Another concern was that of assessment and the Education Review Office reviews (in 1995), including assessment of value-added education for the Education Review Office; feedback by Education Review Office to the Ministry of Education; how to assess for progression in learning and the use of baseline data. The teachers had concerns about the Ministry of Education’s assessment policies, including concerns about performance-based pay; what makes an effective teacher; performance appraisal; putting assessment policy into practice; perceived lack of guidance on putting policy into practice. The teachers had concerns about assessment for summative purposes of the learning related to the science curriculum, including over-assessment; not assessing everything that is taught and learnt; whether to assess all the curriculum objectives or not; assessing to the ‘levels’ in the curriculum; assessment activities to find out what students are thinking and assessing progression.
The notion of quality in educational assessment Another of the international trends in assessment in science classrooms (and in other classrooms) has been research on the development of high-quality assessment procedures (Bell, in press), and is based in the debates on the shift from a paradigm of measurement and psychometric approaches based on true score theory (Black, 2001; Cumming and Maxwell, 1999) to a ‘new paradigm of assessment’ (Gipps, 1994). In educational (including science educational) assessment, quality is not just a technical issue as assessment involves making and acting of choices and judgements, which are underpinned by social values (Berlak et al., 1992; Gitomer and Duschl, 1998; Messick, 1994) and discourses of power (Cherryholmes, 1988). Assessment (of science learning) can be seen as a social practice determined by the specific social, historical and political contexts in
Classroom assessment of science learning
139
which they are undertaken (Gipps, 1999). Given today’s social values, for example on equity, and given the move from psychometric testing and measurement towards educational assessment, quality is no longer thought of in terms of the initial use of the terms of ‘validity’ and ‘reliability’ as previously. The notion of ‘quality’ in educational assessment has been developed to reflect the notions of assessment for educational purposes, that is, formative assessment (Cowie and Bell, 1996b); embedded assessment (Treagust et al., 2001); authentic assessment (Cumming and Maxwell, 1999); holistic assessment (Wiliam, 1994). Rethinking the notion of ‘quality’ has given rise to quality assessment terms such as validity, equity, trustworthiness, fairness (Gipps, 1998); inference, generalisability, consequences, social values (Gitomer and Duschl, 1998); manageability, facility and discrimination (Osborne and Ratcliffe, 2002); reliability, dependability, validity, disclosure, fidelity (Wiliam, 1992); confidence (Black, 1993); and equity, trustworthiness and appropriateness (Cowie and Bell, 1996b). A key indication of quality of educational assessments is that of validity. In the 1970s and 1980s, there was much criticism of the low validity of science assessments for summative purposes used by teachers in science classroom-based assessment (Doran et al., 1993) and in external testing and science examinations, for example for national qualifications (Gauld, 1980; Keeves and Alagumalai, 1998). The meaning of validity expanded as alternatives to pen-and-paper testing were developed (Crooks, Kane and Cohen, 1996). Whereas reliability is affirmed by statistical means, validity relies ‘heavily on human judgment and is therefore harder to carry out, report and defend’ (Crooks et al., 1996: 266). The initial meaning of validity as ‘measuring what it purports to measure’ in relation to traditional (science) multiple choice and pen-and-paper tests has been expanded as the notion of validity has been developed with respect to the quality of alternative assessments, such as performance assessment (Moss, 1992). Crooks et al. (1996) indicate the breadth of current understandings of validity and threats to the validity in their account of eight different stages of the assessment ‘chain’ and the associated threats to validity. For Crooks et al. (1996) the validity of the entire assessment procedure is constrained by the strength of the weakest of the eight links in the validity chain. Hence, researchers have argued that new forms of educational assessment (often called alternative assessments) cannot be fairly appraised unless the older definition of validity is broadened (Linn, Baker and Dunbar, 1991). Research on the broader notion of quality of assessments of science learning include those addressing consequences, equity, fairness, trustworthiness, appropriateness, manageability, fidelity and disclosure. Each of these newer notions of quality is now discussed with respect to the Waikato research on assessment for formative purposes. Consequences Gitomer and Duschl (1998) argued that, typically, the validity of science assessments has been considered only in terms of construct validity – how well the
140
Classroom assessment of science learning
evidence supports the interpretations made on the basis of the assessment. However, Messick (1989) raised the prominence of a second consideration of validity, the consequences of an assessment, that is, consequential validity, which is centrally important to assessment of formative purposes, given that the definition of formative assessment is based on the taking of action to improve learning (Cowie and Bell, 1996b) and that the appraisal is in relation to its effectiveness in improving learning. In considering the concept of consequences, Cowie (2000) asserted that the consequences of formative assessment of science learning – cognitive, social and emotional – cannot be separated out, and ‘therefore adequate and appropriate (valid) ways of generating, interpreting and responding to information gained from students and their learning are those that benefit and not harm student learning, identity, feelings and relationships with others’ (Cowie, 2000: 281). Equity and fairness Equity is an important factor in considering the quality of an assessment and is associated with issues of moral and social justice (Darling-Hammond, 1994) and the equitable and inclusive practice and production of science, multiple world views and science assessment across diversity (Kent, 1996; McKinley, 1997). It implies practices and interpretation of results that are fair and just to all groups, and a definition of science achievement which applies to all students, not just a sub-group (Gipps and Murphy, 1994). In the context of the Learning in Science Project (Assessment) it was used in the sense of providing opportunities for all students to participate in communication and particularly in the science classroom interactions that are the heart of assessment of science learning for formative purposes (Cowie and Bell, 1996b; Torrance, 1993) even if different modes of communication and task formats have to be used (Kent, 1996). Fairness is an aspect of equity and validity. Science students (as with students in general) do not come to school with identical experiences, nor do they have identical experiences at school. Therefore, multiple opportunities for assessment might be needed to provide fairness and comparable treatment for all students in a class – students who will have differing educational experiences – to demonstrate their achievement if they are disadvantaged by any one assessment in a programme (Gipps, 1998). Equity and fairness may be in terms of gender (Gipps and Murphy, 1994) or ethnicity (Darling-Hammond, 1994; Solano-Flores and Nelson-Barber, 2001). The two main messages here are that where differences in science performance are ignored, and not monitored, patterns of inequality will increase and that to ensure assessments of science learning are as fair as possible, we need to address the science curriculum content (the constructs) being taught and assessed, teacher attitudes towards different groups of students, and the assessment mode and item format (Gipps, 1998).
Classroom assessment of science learning
141
Trustworthiness Trustworthiness relates to whether something or someone can be trusted in the classroom setting. It is based on the perceptions of both teachers and students and is an essential element of teaching, learning and assessment, particularly formative assessment (Cowie and Bell, 1996b). Teachers must trust students to provide them with reasonably honest and representative information about their understandings and misunderstandings. Students must trust teachers to provide them with learning opportunities, to show interest in and support for their ideas and questions, to act on what they find out in good faith, and have faith and trust in the assessment practices. Trust in the relationship between a student and teacher in the practice of formative assessment also affects the disclosure by the students of what they know and can do (Cowie, 2000). Cowie stated that from a student perspective, a valid formative assessment is trustworthy; one in which students can have trust in the process as well as the person, where both support and the process do not undermine student understanding, affect and relationships; gives all students access to opportunities to participate in formative assessment; and encourages them to participate in and respond to formative feedback. Appropriateness To be judged appropriate by teachers and students, assessment must be beneficial and not harmful to student learning (Crooks, 1988). Hence, appropriate formative assessment of science learning, for example, is that which is first equitable and trustworthy but also supportive of learning (Black, 1995); indicative of what counts as learning (Crooks, 1988); is matching of the views of teaching and learning used in the classroom (Torrance, 1993); and addresses the importance of science students’ views and the ongoing interactive nature of the practice of assessment of science learning for formative purposes (Bell and Cowie, 1997b). Validity concerns are raised when science students do not give as full a response as they are capable of (Gauld, 1980; Kent, 1996; Shaw, 1997). Manageability An aspect of quality that is of great concern for science teachers is that of manageability – that the assessment is able to be managed within the busyness of the classroom life of teachers and students, and does not detract time from teaching and learning the set science curriculum (Bell and Cowie, 1997b; Stoddart et al., 2000). Fidelity and disclosure Wiliam (1992) identified two issues, disclosure and fidelity, which may limit the information teachers have to notice and recognise in interactive formative
142
Classroom assessment of science learning
assessment (of science learning). The disclosure of an assessment strategy is the extent to which it produces evidence of attainment (or non-attainment) from an individual in the area being assessed (Cowie and Bell, 1996b). Wiliam (1992) defined fidelity as the extent to which evidence of attainment, which has been disclosed, is observed faithfully. He claimed that fidelity is undermined if evidence of attainment is disclosed but not observed. For example, the teacher may not hear a small-group discussion in which the science students demonstrate they understand a scientific concept. Fidelity is also undermined if the evidence is observed but incorrectly interpreted, for example, if the teacher did not understand the student’s thinking in a science lesson – a possibility if there is insufficient commonalty in the teacher and the students’ thinking.
Theorising of assessment Another trend in research and development in assessment in science classrooms (as in other classrooms) has been in the theorising of assessment practices (Bell, in press). The Waikato research on assessment for formative purposes has contributed to this theorising. If teaching, learning, assessment and curriculum are considered in an integrated and interdependent way (Carr et al., 2000), one might theorise them in a similar way. As theorising about learning and teaching has developed from a behaviourist to cognitive science to sociocultural views (Barker, 2001) so too has theorising of assessment. For example, there has been theorising of assessment to be consistent with a constructivist view of learning (Berlak et al., 1992; Gipps, 1994a; Wiliam, 1994) and theorising of assessment in general as a sociocultural practice (Broadfoot, 1996; Filer, 2000; Filer and Pollard, 2000; Gipps, 1999; Pryor and Torrance, 2000). In the more recent Waikato research, assessment of science learning for formative purposes has been theorised as a sociocultural practice (Bell and Cowie, 2001b) and as a discursive practice (Bell, 2000), which is also evident in the science assessment literature (Fusco and Barton, 2001) and the general assessment literature (Sarf, 1998; Torrance, 2000). To view assessment as a sociocultural practice is to view it as value-laden, socially constructed and historically, socially and politically situated. That is, one can never do assessment separate from one’s own history (individual or social) or outside its contexts. As Gipps (1999) said, ‘to see assessment as a scientific, objective activity is mistaken; assessment is not an exact science’ (p. 370). Assessment of science learning may be viewed as a purposeful, intentional, responsive activity involving meaning making and giving feedback to students and teachers, to improve learning; an integral part of teaching and learning; a situated and contextualised activity; a partnership between teacher and students; and involving the use of language to communicate meaning (Bell and Cowie, 2001b). Theorising assessment as a sociocultural practice raises several issues for researchers. One is the issue of ‘whose theorising’ and the purpose of the theorising. In the Waikato research of assessment of science learning (Bell and Cowie,
Classroom assessment of science learning
143
1999) and in a UK study of classroom formative assessment in general ( Torrance and Pryor, 2001), the teachers and university researchers involved have been encouraged to theorise their own assessment practices and to develop classroom practice models, using their own shared vocabulary. This theorising was identified by the teachers as an important aspect of their teacher development practices. A related issue is that of the unit of analysis which is ‘the event, rather than the individual is the primary unit of analysis for evaluating learning environments from a socio-cultural perspective . . . The key issue in studying innovative curricula is the knowledge practices in which learners collectively participate’ (Hickey and Zuiker, 2003: 548). This is evident in the use of cameos of assessment of science learning (Bell and Cowie, 1997b) and ‘incidents’ in research on formative assessment in general (Torrance and Pryor, 1998). If assessment of science learning is theorised in terms of a sociocultural view of mind, the implications (Gipps, 1999) include that: assessment can only be fully understood if the social, cultural and political contexts in the classroom are taken into account; the practices of assessment reflect the values and culture of the classroom, and in particular those of the teacher; assessment is a social practice, constructed within social and cultural norms of the classroom; what is assessed is what is socially and culturally valued; the cultural and social knowledge of the teacher and students will mediate their responses to assessment; assessments are value-laden and socially constructed; a distinction needs to be made between what a student can typically do (without mediational tools) and best performance (with the use of mediational tools); assessments need to give feedback to students on the assessment process itself to enable them to do self and peer formative assessment; and teachers and students need to negotiate the process of assessment to be used, the criteria for achievement, and what counts as acceptable knowledge. In summary, the Waikato research on assessment has contributed to national and international debates on the political contexts of assessment, the multiple purposes for assessment, the assessment of multiple goals of science learning, assessment for formative purposes, assessment for summative purposes, the notion of quality in educational assessment, and theorising of assessment (Bell, in press).
144 Culture,7M1ori and science education Chapter
Culture, M2ori and science education The social, cultural, political and historical contexts
Introduction In considering learning, teaching and assessment in science education, a sociocultural view requires us to consider not only the individual, but also the social, cultural, historical and political contexts in which the teaching, learning and assessment is occurring. In such a view, the teaching, learning and assessment of science are viewed as social and cultural practices (Bell, 2000). Culture cannot be removed from accounts of learning, teaching and assessment of science. In this chapter the research and debates on culture and science education in New Zealand are reviewed. Although the research reviewed in this chapter is necessarily situated in the social, cultural, historical, political context of New Zealand, and is addressing the science education of New Zealand’s indigenous people, Maori, it will be of interest to all science educators concerned with addressing culture and science education, whether it be in contexts such as urban science education in the USA (Calabrese-Barton and Tobin, 2001); multicultural science education in the UK (Ditchfield, 1987; Hodson, 1993); aboriginal science education in Australia (Christie, 1990); antiracist science education (Gill and Levidow, 1987); the science education of First Nation peoples in Canada (Aikenhead, 1997); science education and Pasefika peoples (Moli, 1993b); Muslim science education (Loo, 2001); or the science education of students in Africa ( Jegede, 1995). Whilst aspects may differ due to the New Zealand social, cultural, political and historical situation, there are some similarities that will be informative to the international community of science education researchers. This chapter and Chapter 8 review the Waikato research, national curriculum developments and debates on culture and science education in New Zealand. But these are not abstract, academic debates; they are driven by a strong commitment by Maori to improve the learning outcomes in science education for Maori students, so that they are educated to take advantage of what science has to offer – in everyday life and the world of work. The New Zealand debates on culture and science education are grounded in the considerable concern for the underachievement of Maori students in science education:
Culture, M 1ori and science education
145
Maori children do not perform well in science. While the highest achieving Maori children match their Pakeha and Asian counterparts, more than three quarters of Maori children score below the means for these groups. (McKinley, 2000: 33) In documenting and discussing the research in Maori science education at Waikato, I first position myself in the political debates as a Pakeha. (Pakeha is a Maori word for New Zealanders of European descent. A glossary of Maori words is given at the end of the book.) I then give an account of the social, cultural, political and historical contexts of the work at Waikato using the work of Maori colleagues, particularly colleagues at Waikato. A discussion of culture and science education cannot be discussed without these contexts, which are an integral part of the research and theorising.
A historical and political background: one P2keh2’s perspective I am a fifth generation-born New Zealander as both my mother’s and father’s family migrated to New Zealand in the first wave of migration, settlement and colonisation from the UK, in the 1840s. My maternal great-great-great grandmother, Elizabeth Avery, her husband and eight children emigrated to New Zealand on the Bolton, under the auspices of the New Zealand Company, leaving Gravesend, England, in November 1839, arriving at Port Nicholson (Wellington, New Zealand) in April, 1840 (King, 1991). My paternal great-greatgreat-grandfather, John Bell, a shoemaker from Gillingham, Dorset, England, emigrated with his wife, Jane and family to New Zealand on the SS Birman, arriving in Wellington in 1842. But despite my families’ long history with New Zealand, I myself had very little or no contact with Maori until I was a teacher. At school, I learnt a curriculum about things Maori before European settlement in art and social studies lessons. Outside the education system, I learnt a curriculum of contemporary Maori in my family gatherings at Christmas in the 1950s and 1960s. My wider family ‘taught’ me about the ‘inferior nature of Maori’ – the commonly held racist view of the time. In listening to my uncles and aunts tell of their childhoods and school days, I learnt that Maori were ‘poor, bad farmers, unintelligent, lazy, stole school lunches’ and had killed Pakeha in the Land Wars/New Zealand Wars. My family’s view of Maori was a product of the times and was no different from the commonly held view by many Pakeha at that time. As my education contained no Maori knowledge and narratives of New Zealand after the signing of the Treaty of Waitangi, and I had not met any Maori, I had no knowledge or experiences to counteract these strong messages. My first experiences and knowledge of Maori were to be gained mainly through my professional life as a pre-service teacher, teacher and educational researcher in New Zealand.
146
Culture, M1ori and science education
Hence, although I was well educated, having received a primary, secondary and tertiary education and a professional teaching qualification in New Zealand in the 1950s–1980s, my education and curriculum did not reflect the bicultural nature of our nation. It had contained little if any contemporary Maori culture and language, nor had it included a history of my own country as told by Maori. I had not been educated to be a bicultural citizen of New Zealand. And what I had learnt from out-of-school sources (television, books, family conversation) was founded on racism and oppression. Over the years, as a person growing up and working in New Zealand and working overseas, I have learnt that Maori are different from my family, and in ways that I did not fully understand, and that by the mid-1980s I was different from many in my wider family in my views of Maori. My story is similar to that of many of my peers who also undertake research, curriculum development and policy analysis in science education in New Zealand today. From the mid-1980s onwards, what I have learnt is a story of Maori resistance and activism against the educational and other policies of the dominant culture. While British and other western cultures have brought advantages for Maori, state education has also been a story of institutional and epistemological racism for Maori. I now document this story based on the writings of Maori colleagues.
A historical and political background: a M2ori perspective New Zealand is a multicultural society which includes Maori, Pakeha (people of European descent), Pacific, Asian and African peoples. However, New Zealand is a bicultural nation as the Treaty of Waitangi, the founding document of this nation, was signed on 6 February 1840 by two peoples: Lieutenant-Governor Hobson on behalf of the British Crown and 512 chiefs of the Maori people. Pakeha settlers had no involvement in signing the Treaty. At the time of signing the Treaty, there were about 110,000 Maori and 2000 white settlers, missionaries and whalers (King, 2003). (In 2004, approximately 600,000 (15 per cent) of the total population of 4 million identify themselves as Maori.) In terms of social policy and education, the Treaty guaranteed to Maori, under the three main articles, partnership – a share in the power of decision making; protection – the power to define and protect knowledge and language; and participation – a share in the benefits of participation in education (Bishop and Glynn, 1999). The Treaty guaranteed to Pakeha the right to settle and live in Aotearoa New Zealand. Therefore, New Zealand as a modern independent state was not founded by colonial conquest or the illegitimate invasion of settlers but on the signing of a treaty (Wilson, 2002). But: The development of New Zealand since the signing of the Treaty of Waitangi in 1840, despite continual armed and passive resistance by Maori people,
Culture, M 1ori and science education
147
has been one where the Pakeha majority has benefited enormously and where Maori have been politically marginalised, culturally and racially attacked, and economically impoverished within their own country. (Bishop and Glynn, 1999: 14–15) Colonisation created power imbalances in New Zealand (Bishop and Glynn, 1999: 53), resulting in: • •
•
•
the perpetuation of a belief in cultural superiority which promotes feelings of superiority in some and inferiority in others; the perpetuation of an ideology of ‘we are all New Zealanders’, therefore legitimating the belief that all children are the same and need the same treatment; those who fail deserve to because of some inherent personal problem or cultural deficiency . . . ; the denial that alternative world views exist with differing cultural aspirations, preferences and practices; this perpetuates the idea of normality being based on conformity rather than diversity . . . ; the perpetuation of deficit explanations, with the consequent impact on the victims of these. (Bishop and Glynn, 1999: 53–4)
Up to the 1970s, successive governments and Department of Education policies of assimilation and integration have perpetuated this discourse in education, despite the intention to address Maori underachievement in schools (Walker, 1991). These two policies are now discussed in more detail as they have influenced, in some way, the thinking of people involved in science education in New Zealand today. At the beginning of the bicultural nation of New Zealand in the 1840s (less than 170 years ago), government and educational policies were one of assimilation: ‘a policy that was based on a racist assumption; namely, that which had been brought by the colonists was the best for Maori people. Maori were encouraged to abandon their culture as rapidly as possible in order to learn the ways and processes of the dominant culture’ (Bishop and Glynn, 1999: 16). This policy of assimilation was to be official government policy until the 1960s and the practice of some teachers for many years after that. Maori culture was seen as inferior: less intelligent, less civilised, less moral, less well developed as it did not then have a full (that is, written) literature, and was seen as unable to cope with complex human problems. Maori were forced to abandon their culture by the assimilationist policies and practices of conversion to Christianity; oppression of the use of the Maori language; a school curriculum restricted to manual and technical instruction; the simplification of Maori cultural knowledge, and the promotion of the idea of the inability of Maori cultural knowledge to cope with complex human problems, such as celestial navigation and commerce; and school textbooks which deliberately demeaned, marginalised and vilified Maori
148
Culture, M1ori and science education
culture, denying Maori children their culture and language. Both Maori and Pakeha school children read these textbooks. Bishop and Glynn (1999) used the term ‘epistemological racism’ after Scheurich and Young (1997), meaning racism that is embedded in the fundamental principles of the dominant culture, to describe these assimilationist policies and their effects. The result is the marginalisation of Maori cultural aspirations, preferences and practices in the education system and the continuance of Maori ‘underachievement’ in a system that was designed to promote such underachievement (Bishop and Glynn, 1999: 13). Bishop and Glynn (1999) further assert that failure of Maori in schools is inevitable if the objective is assimilation or the perpetuation of a situation of dominance and subjection. It is argued that these assimilationist policies and their inherent racism are still evident in discourses in education today (McKinley, 1995b, 2003). And in the 1950s, it was these policies that kept Maori cultural knowledge and narratives out of the school curriculum for Maori and Pakeha students. The migration of Maori to the cities and towns in the late 1940s, 1950s, and 1960s brought the problem of Maori education out into the open. In the 1960s, the official policy on Maori education (and the address to the challenge of cultural diversity) became one of integration, that is, the Maori culture and language was to be incorporated and integrated into the dominant Pakeha culture. So ‘instead of the culture and language of the, by now, numeric minority being destroyed, all minority groups were to be integrated into the culture of the dominant group. In effect, it was suggested that by this means the best of both cultures would be integrated into one culture’ (Bishop and Glynn, 1999: 37). One key result of the educational policies based on integration was the development of cultural deprivation theories or deficit theories, which see Maori children as ‘culturally deprived’, to explain the underachievement and failure of Maori children at school. The cultural deficit theory and integration policies: ‘proposed . . . a further denial of Maori culture and language as a means of addressing life’s problems. Such an approach reinforced the notion that living as Maori posed a problem, or barrier to learning – in short, a cultural deficiency’ (Bishop and Glynn, 1999: 37). Empirical educational researchers used this theory of ‘cultural deficiency’ not only to explain their data on the underachievement of Maori students but to determine the research questions to be investigated, namely whether or not there were any significant differences between Maori and European schoolchildren on tests of scholastic achievement and certain selected determiners, i.e. intelligence, home background, attitude to school, speed of performance and listening comprehension, for example (Lovegrove, 1966). This research promoted the discourse of the ‘deprived nature of Maori homes’ as well as that of Maori homes having a deficient cultural background. Maori homes, from this perspective, therefore created a context where children were ‘suffering a pathology which manifested itself in inadequate language and intellectual development. Such pathologies were the result of a “deficit cultural background” ’
Culture, M 1ori and science education
149
(Bishop and Glynn, 1999: 38). ‘Nevertheless, this focus on cultural deficiencies (rather than racial inferiority) itself continued to engender some depressing conclusions about the state of Maori education over a decade later’ (Bishop and Glynn, 1999: 39). This was due to the implication that achievement differences could not be ameliorated by changes to the educational system and a questioning of whether the equalising of performance should be a goal. By the 1960s, it appeared that policy makers saw the Maori culture, if not the language, as having a role in New Zealand schools. The language was presumed to be dying a natural death as a community language but would probably be kept for ceremonial occasions such as on the Marae. Resources covering Maori art and craft, history, music and games began to emerge in greater numbers. (McKinley, 1995b: 28) But what was taught was restricted to pre-colonisation times and little was taught of contemporary Maori cultural knowledge: ‘any Maori language being taught was done in much the same manner as any other foreign language’ (McKinley, 1995b: 28). Again, the integration policies kept much contemporary Maori knowledge from the school curriculum in the 1960–1970s for both Maori and Pakeha students. It was within these social and cultural discourses that the research in science education at Waikato began in the 1970s. In the twenty-year time frame of this review (1979–98), there were four main masters or doctoral research investigations into Maori and science education (Gribble, 1993; Kent, 1996; McKinley, 1995b; Stead, 1984).
The Early Waikato Research The research of the first Learning in Science Project was undertaken in the late 1970s/early 1980s, when the discourses of assimilation, integration and cultural deficit theory were in use. In addition, the first Learning in Science Project team was comprised of Pakeha who, like me, had been educated at the time of government policies of assimilation and integration, and had come from families, like many Pakeha families, with varying attitudes towards Maori. As a part of the first Learning in Science Project, Keith Stead accepted the task of researching students’ outlooks on science (the third strand of the research project) (Stead, 1982) and in particular, for his doctoral research (Stead, 1984), the reasons and explanations for the under-representation of Maori (as well as Pacific Island students and girls) in science education. In his research, Stead used the term ‘children’s outlook on science’ to encompass students’ affective response to science, including ‘such dimensions as interest, motivation, aspirations, beliefs, expectations and values, pertaining to science’ (Stead, 1984: 11). Hence, within the general term ‘scientific attitudes’, he
150
Culture, M1ori and science education
distinguished between scientific habits and attitudes to science. Furthermore, he distinguished within the notion of ‘outlook on science’ two main components: the cognitive and conative. The cognitive consisted of the conceptual (both ability and style) and the perceptual (both objective and subjective). The conative consisted of the attitudinal and social (Stead, 1984: 233). Stead considered that a complete explanation of the different outlooks on science held by the different groups of students required both a consideration of quantitative test results and qualitative data collected as part of his research. The quantitative test data indicated that European, Maori and Pacific Island students brought different cognitive abilities to the study of science (Stead, 1984: 268). Differences were found between European and Maori on all cognitive measures – a Science Questionnaire; TOSCA scores; Cognitive Style measures – the field independence/dependence scores as measured by the Hidden Figures test; relational/categorical conceptual styles as measured by the Conceptual Style test; locus of control measures as measured by the Test of Intellectual Achievement Responsibility test. This research theme was typical of educational research (by Pakeha) in New Zealand in the 1960s and 1970s (Bishop and Glynn, 1999). Stead explained these differences, not in terms of innate racial intellectual capacity but in terms of the cultural experiences, knowledge and values promoted in the school system and used as criteria of success. The qualitative data were obtained from students using conversations based around the repertory grid technique, after Kelly (1955). The parents/caregivers of the students were also interviewed using an unstructured interview. Questionnaires using Likert scale responses were also devised to measure the conative aspect of children’s outlook on science. The qualitative data was used to provide insights with which to explain the quantitative test results. The qualitative data analysis indicated that Polynesian (Maori and Pacific Island) students perceived there to be ethnic role stereotyping in science education: (a) many Polynesian (particularly the Maori) students did not have high academic self-esteem and so tended to reject much of the school’s curricula for not offering sufficient of personal value to them. Therefore, Science, in being perceived by them as an example of an academic subject of low social relevancy, was so ‘rejected’. (b) School was perceived primarily as preparation for work where, for many Polynesian students, social affiliations were held to be of high priority. Consequently, the limited stereotype the majority of students held of a ‘scientist’ (that of a white-coated, Caucasian male, working in apparent isolation from other people) precluded close identification with such a vocational role . . . because many Polynesian students perceived science not to be particularly concerned with people (but with formulae, abstract principles, and ‘things’), it was perceived as being ‘unimportant’ . . . (Stead, 1984: 165–6)
Culture, M 1ori and science education
151
Stead theorised the under-representation in terms of the ‘sociocultural’, for example the status of schooling in the students’ value systems. In summarising the findings of his qualitative and quantitative data, Stead stated that ‘the ethnic differences in students’ outlooks on science could be accounted for by the “cultural inequality” ’ present in New Zealand society (Stead, 1984: iii). He considered that the cultural inequality that existed in the New Zealand educational system found expression in the manner in which success and failure are defined in ways commensurate with European achievement criteria. Maori cultural knowledge was not valued and hence its possession by students did not give them measures of success. Success was only achieved if Pakeha knowledge had been obtained. Stead’s research is notable for several reasons: • •
•
•
it was one of the first pieces of research into Maori and science education; the first Learning in Science Project team was concerned about the underrepresentation of girls and Maori and Pacific students, but the findings of Stead’s research did not inform other strands of the first project or the following Learning in Science Projects; it was a product of its time for it included two approaches to the question of science education for Maori students, mirroring the two approaches to Maori education at that time. The quantitative data analysis was essentially an examination of the differences in the test scores of Maori, Pacific and Pakeha students. The test instruments were examined for cultural bias and the test score differences explained. This aspect of the research primarily met the agenda of the funding agency (the New Zealand Council for Educational Research) rather than those of the Learning in Science Project. But it did serve to link Stead’s research with the then current research into Maori education in the early 1980s. Stead also used the category of ‘Polynesian’, indicating that Maori and Pacific students could be perceived as a homogeneous group for the purposes of data analysis, in much the same way as the category ‘women’ was used as a homogeneous group. Whilst a common view then, it is not today; the qualitative work was more in keeping with the aims of the Learning in Science Project and is of interest in that the views of Maori parents and academics were sought as to the under-representation of Maori in science education. In addition, some of the numerous reports on Maori education, in which Maori voices were embedded, were also examined.
Stead included the following quotations in his thesis. The first one, from a report by the National Advisory Committee on Maori Education, indicated the growing impatience of Maori at the government’s lack of effective action: The time for turning a blind eye to Maori underachievement in education is past – time is no longer on our side, and we must act urgently. New Zealand will not achieve social cohesion and enhanced mutual respect between the races until Maoris are represented in due proportions in all walks of life,
152
Culture, M1ori and science education
taking their place in the prestige occupations (page 6) . . . It is clear that if we are to meet the educational needs of Maoris our educational system requires improvement. The community and teachers alike need to make sure that the system is not simply striving to do the wrong things more efficiently. We need to examine closely our educational philosophy, and our classroom strategies, the organisation and climate of our schools, and where necessary, change them (page 7). (National Advisory Committee on Maori Education, 1980) The second one referred to the assimilationist and integrationist policies of the government as far as Maori education is concerned: Look at the question, education for what? What are you trying to educate us for? To be better road sweepers? To be better freezing workers? We don’t need an education for that. If you are educating us to be better teachers, to be better accountants, to be better scientists, then what you really want us to do is to become Pakehas, and we can’t be Pakehas and stay ourselves in the community. But, that’s where we find our greatest satisfaction and that is our means of success. (Mahuta, 1980, quoted in Stead, 1984) In hindsight, the major critique of this work is that this research, like much educational research of the day, did not address the political aspects of power and control in the classroom and New Zealand society (Bishop and Glynn, 1999; McKinley, 1995b, 2000), the term ‘cultural inequality’ being used instead. The main benefit of this research was for Pakeha, in educating us about how Maori perceived themselves with respect to science education. Maori are the most researched group in educational research yet the large volume of reports and research on Maori and education have yielded little improvement to the situation for Maori. In the early 1980s, Maori were, yet again, stating to the government that they had solutions to the problems and that they needed the funding for these initiatives. They did not want the funding given to Pakeha-initiated solutions that raised the awareness of Pakeha on things Maori but which made no difference in improving educational outcomes for Maori (Bishop and Glynn, 1999). The government’s insistence on funding Pakeha-initiated solutions for Maori underachievement perpetuated the racist view that Maori are not capable of making educational policy for their own children. Maori rightly claimed that they had solutions that would work: The logical outcome of this argument is that the ‘problem’ is somehow located in being Maori. At the same time, many Maori educational researchers . . . have also stated that part of the solutions to Maori educational underachievement and participation also is located in being Maori. (McKinley, 1995b: 3)
Culture, M 1ori and science education
153
National curriculum developments in the 1980s There was no further research on Maori and science education at Waikato during the rest of the 1980s. It was not an aspect of the two Learning in Science Projects on primary science education or on the teaching and learning of the concept of ‘energy’. What is of interest is that in the second half of the 1980s, the ideas of the Learning in Science Projects, although not specifically the findings of Stead (1984), were taken into account in national science curriculum development. It was also during this time that Maori views on education for Maori were being addressed in the developments of multiculturalism, biculturalism, bilingualism, total immersion and kura kaupapa Maori. These were to have an influence on the national science curriculum developments, which in turn influenced the research at Waikato in the 1990s on Maori and science education. These Maori initiatives and the national science curriculum developments of the 1990s are discussed in this section. Taha M3ori and the science curriculum In the mid-1980s, staff of the Curriculum Development Division, of the former Department of Education, were expected to initiate curriculum development work based on Taha Maori policy. The Taha Maori (literally ‘the Maori side’, pertaining to Maori) programme (Smith, 1990; Walker, 1985) in schools was a bicultural curriculum development, through which multiculturalism would be achieved (Metge, 1980). This programme was an initiative of the former Department of Education in the early 1980s in response to the growing call among Maori and non-Maori educators for some recognition of the place of Maori as tangata whenua (indigenous people) of Aotearoa (the Maori name for New Zealand). Taha Maori was defined as ‘the inclusion of aspects of Maori language and culture in the philosophy, the organization and the content of the school’ (Department of Education, 1984: 1). The Taha Maori policy had two aims: one was that Taha Maori programmes would validate Maori culture and language in the minds of Pakeha New Zealanders. This aim was largely achieved and resulted in Maori content being included in the curriculum for primary and secondary students. Since the signing of the Treaty 140 years earlier, Maori knowledge was being included in the school curriculum for the first time. The second aim was to raise the underachievement of Maori students by raising their sense of self-worth and sense of identity. This second aim was not met (Bishop and Glynn, 1999), as the curriculum remained geared to the needs and aspirations of the members of the majority culture. As a leading Maori educationist said: ‘The curriculum initiative of Taha Maori is a Pakeha-defined, initiated and controlled policy which serves the needs and interests of Pakeha people’ (Smith, 1990: 183). Bishop and Glynn, 1999 stated:
154
Culture, M1ori and science education
Even if Taha Maori was originally intended to address issues of Maori achievement in schools, it is most unlikely that it could have addressed the underlying issue of retrieving a colonised language and culture because Taha Maori (as an implementation of biculturalism) was designed to met the aspirations of the Pakeha majority and not those of Maori people. Attempts to promote biculturalism within mainstream education has meant that the focus of change has been on the Pakeha society. This focus on teaching Pakeha people about Maori culture has often consumed resources at the expense of Maori language and cultural aspirations. (Bishop and Glynn, 1999: 42–3) In science education, Taha Maori policy was implemented through the use of Maori contexts and examples in the national curriculum review and development of the then Years 7–11 (Forms 1–5) science syllabus (Bell, 1987). The use of Maori contexts was supported by the Waikato research on constructivist views of learning, including the context and learning experiences that were meaningful and relevant for the students, the existing knowledge and experiences that Maori students brought to the lesson and helping students make links between their existing ideas and the science ideas being taught (Bell, 1991a). At the time of the first Learning in Science Project, with its personal constructivist theorising, existing knowledge was not seen as including cultural knowledge. This connection was, however, made during these science curriculum developments of the late 1980s. There were criticisms by Maori of the way Taha Maori was implemented in the science curriculum, the key one being that the only Maori knowledge and language that would be accepted into science teaching was knowledge that would fit under the definition of what could be considered as science. Hence, looking at making hangi (cooking in the ground using heated stones and steam) was acceptable as it could be explained in terms of heat capacity – the real science knowledge. (McKinley, 1995b: 33) Maori cultural knowledge about hangi was not considered to be important in itself, but as a vehicle to learning the ‘real’ science. Te reo M3ori and the science curriculum At the same time as the Taha Maori developments, in the early 1980s, there were the Maori-initiated developments of bilingual (English and Maori), total immersion (only Maori language used) and kaupapa Maori (structured around the use of Maori philosophy) classes/units and schools. These alternatives to the mainstream schools were to raise the achievement of Maori students through the
Culture, M 1ori and science education
155
use of Te Reo Maori as the language of instruction. Language is seen as central to the maintenance of Maori culture and identity (May, 2000) and there were concerns expressed at the decline in the number of fluent Maori speakers (Benton, 1979). Before the 1890s, the majority of Maori could speak their indigenous language but colonisation and consequent development of the nation convinced many Maori to place more importance on the learning and speaking of English. By the 1970s, there were two or three generations of Maori who spoke only English. The loss of language (and culture) was of concern as it was linked, by Maori, to a low self-esteem in Maori. This was and is still reflected in the statistics on crime, health, unemployment and educational achievement (Bishop and Glynn, 1999). ‘Maori language is used to validate Maori identity by confirming, legitimising and reproducing Maori cultural values, beliefs, resources and practices’ (McKinley, 1995b: 43). ‘The loss of the language was also a concern in that, unlike other cultural groups in New Zealand, there was no other place in the world where the language would be preserved and developed . . . (McKinley, McPherson Waiti and Bell, 1992: 580). In 1986, the New Zealand Waitangi Tribunal announced its finding that the education system in New Zealand was operating unsuccessfully because too many Maori children were not reaching an acceptable standard of education and the educational system as it operated was in direct contravention of the Treaty of Waitangi. The outcome from this claim was the passing into law of the Maori Language Act in 1987. The Act contains three significant provisions: 1 2 3
Maori was declared to be an official language of Aotearoa New Zealand (along with English). Maori was permitted in legal proceedings. Te Taura Whiri i te Reo Maori (the Maori Language Commission) was established.
Te reo Maori could now be used in the classroom, the law courts, parliament, in addressing mail, in commerce, for example writing a cheque, and in the media. Since becoming an official language, some Maori words and phrases have become part of everyday New Zealand speech to the extent that they are not translated, for example, tangi (funeral), mana (prestige), kuia (respected woman elder), koha (loosely meaning a donation), kaimoana (seafood), kia ora (a greeting), tapu (sacred). In addition, Maori cultural practices were becoming accepted as a part of Pakeha cultural practices, for example at funerals, the open casket for viewing before the service, and the open invitation to everyone to make a contribution to the eulogy. But the acceptance was not always plain sailing. In the mid-1980s there was a public and media debate and controversy about the propriety of a public service telephone operator using the Maori greeting ‘kia ora’ when answering a phone call to a government department. There was a similar public and media debate when the New Zealand national anthem was
156
Culture, M1ori and science education
sung in Maori only, when the All Blacks (the New Zealand national rugby football team) played a test match against England, at Twickenham, England, in October 1999. The findings of the Learning in Science Projects were not directly used to theorise the teaching of science in Maori. Whilst the project findings did acknowledge the interrelationships between language and cognition (Bell, 1981b; Bell and Freyberg, 1985), the main arguments for the teaching of science in te reo were political and about social justice; the honouring of the Treaty of Waitangi and what it means to be a Treaty (of Waitangi) partner (Bell, 1990). As May (2000) stated: language loss is not only, perhaps not evenly primarily, a lingusitic issue – it has much more to do with power, prejudice, (unequal) competition, and in many cases, overt discrimination and subordination . . . Thus, it should come as no surprise that the vast majority of today’s threatened languages are spoken by socially and politically marginalised and/or subordinated national minority and ethnic groups. (May, 2000: 368) The first bilingual school was set up in Ruatoki in 1978 and the first secondary bilingual unit was established at Wellington High School in 1982. The rationale for the establishment of bilingual education according to Spolsky (1987) included the revitalisation of the Maori language; provision of equitable educational opportunities for Maori students; establishment of appropriate venues for the transmission of spiritual and cultural values to Maori children; and a mechanism whereby Maori are able to exercise greater control over future directions of Maori education (McKinley, 1995b). There were criticisms of the bilingual programmes such as resourcing; the tensions between academic and ethnic aims; and the lack of trained teachers who are both fluent in te reo Maori and qualified in other subject areas (Barton, 1990; Ohia, 1990). In response to these criticisms, total immersion programmes were developed by Maori as another attempt to get the mainstream education system to cater for Maori needs. Total immersion schools developed from mainstream English-only schools, through a bilingual phase. However, the use of the Maori language as the medium of instruction did not necessarily structure the curriculum around Maori values and cultural knowledge. At times, the English language was translated and the Pakeha knowledge and values maintained. In addition, the government was slow to fund the bilingual and total immersion units/schools. The debate and the Maori initiatives moved to the setting up of state-funded alternatives to the mainstream based on Kaupapa Maori (Maori philosophy and principles): kohanga reo (Maori medium pre-schools), kura kaupapa Maori (Maori medium primary schools), whare kura Maori (Maori medium secondary schools) and whare wananga Maori (Maori tertiary institutions).
Culture, M 1ori and science education
157
As Liz McKinley stated: ‘Perhaps the most influential in shaping the issues movements of the 1980s was Maori sovereignty . . . ‘Increasingly over the last 20 years, Maori have demanded Maori control of Maori education because they have become increasingly frustrated at the inability of the mainstream education system to respond fast enough to their needs . . .’ (McKinley, 1995b: 38). Graham Smith (1997) stated that particularly since the advent of kohanga reo (total immersion early childhood centres) in 1982, Kaupapa Maori has become ‘an influential and coherent philosophy and practice for Maori conscientisation, resistance and transformative praxis to advance Maori cultural capital and learning outcomes within education and schooling’ (p. 453). Kaupapa Maori was a Maori initiative and response to the failure of a succession of government policy initiatives of assimilation, integration, multiculturalism and biculturalism to sustain Maori educational, cultural and language aspirations. Kaupapa Maori initiatives are epistemologically based within Maori cultural specificities, preferences and practices and are a ‘means of proactively promoting a Maori world-view as legitimate, authoritative and valid in relationship to other cultures in New Zealand’ (Bishop and Glynn, 1999: 65). The first kura kaupapa Maori was established in 1985 in Auckland without government funding, but since 1989 government funding has been available. At the end of 2003, there were 61 kura kaupapa Maori and close to 6000 students are enrolled in kura kaupapa Maori (Ministry of Education, 2003a). The kaupapa Maori principles (Smith, 1997) include: 1
2
3
4
Tino rangatiratanga (the ‘self-determination’ principle) The goal of this principle is control over one’s own life and cultural well-being and has been facilitated by the autonomous structure of kura in that the organisers of the Kura have to make all the necessary decisions about administration, staffing and management. Taonga tuku iho (the ‘cultural aspirations’ principle) In the kura kaupapa Maori, to be Maori is to be normal, taken for granted. Maori language, knowledge, culture and values are valid and legitimate. Maori cultural aspirations are answered and a strong spiritual and emotional factor is involved to evoke and support the commitment of Maori people to the initiative. Ako Maori (the ‘culturally preferred pedagogy’ principle) The goal of this principle is culturally appropriate teaching and learning settings and practices, which are able to connect closely and effectively with the diverse cultural backgrounds and life circumstances (socio-economic) of Maori communities. Other cultural pedagogies may be used, but only on the choice of Maori and only if the Maori language, knowledge and cultural values are not undermined. Kia piki ake i nga raruraru o te kainga (the ‘socio-economic’ mediation principle)
158
5
6
Culture, M1ori and science education
This principle refers to the way participation in kura reaches into Maori homes and brings parents and families into the activities of the school. Whanau (the ‘extended family structure’ principle) Whanau is a primary cultural structure that allows for Maori cultural practices, values and thinking. It is an extended family social structure in which the group takes responsibility to assist and intervene in problems, rather than individuals. It also functions to support Maori parents who had had a hostile schooling, to reinvest in the schooling and education of their children. Kaupapa (the ‘collective philosophy’ principle) Kura Kaupapa Maori have as a collective vision a charter called Te Aho Matua (the saying of the parents) which provides guidelines for what constitutes excellence in Maori education. Further, ‘its power is in its ability to articulate and connect with Maori aspirations, politically, socially, economically and culturally’ (Smith 1997: 472).
Maori see that Maori language, Maori education and Maori development are inextricably linked. They see language as absolutely essential to their essence, their well-being and their identity as Maori. Maori people want to maintain their integrity in terms of te reo (language), tikanga Maori (Maori customs) and a matauranga (knowledge) base. That base is the framework from which Maori people are able to express themselves and participate in the world. (Bishop and Glynn, 1999: 97) The current situation of Te Reo Maori can be expressed in the following statistics. In 2001, the Statistics New Zealand survey on the ‘Health of the Maori language’ (Statistics New Zealand, 2001) indicated that whilst most Maori speak English, only 42 per cent of Maori aged 15 and over (about 136,700 people) have some Maori language skills. Of these, 9 per cent can speak Maori ‘very well’; 33 per cent could speak Maori ‘fairly well’ or ‘not very well’. Therefore, 58 per cent of Maori surveyed could speak ‘no more than a few words or phrases’. The ability to carry out a conversation in Maori increased with age in the survey results. The census data, to which the survey was connected, also indicated that Maori is the most widely spoken second language in New Zealand (Maori Language Commission, 2002). At July 2003 there were 5793 Maori students enrolled in 61 kura kaupapa Maori throughout the country. Also at July 2003, Maori language was being learnt by 26.5 per cent of all Maori students through immersion or bilingual programmes or as a school subject (Ministry of Education, 2003a). But in the 1980s there were few published school resources for teaching science (and other subject areas) in Te Reo Maori.
Culture, M 1ori and science education
159
M3ori and the science curriculum The two developments of Taha Maori in the science curriculum, and the teaching and learning of science in te reo Maori were included in the 1980s Curriculum Review in Science (CRIS). This curriculum development, essentially to revise the Years 7–11 science curriculum (Bell, 1990, 1991a), drew on the notion of ‘science for all’ (Fensham, 1985). Here ‘science for all’ was interpreted as inclusive: that western science was for all students, not just for Pakeha students but also for Maori students, who had a history of underachievement and/or under-representation in science in schools. The interpretation of ‘science-for-all’ as science–technology–society (STS) has not been as visible in New Zealand as it has in other countries, for example in the USA (Yager and Tamir, 1993). Instead, the New Zealand interpretation focused on the inclusion of Maori (and girls) in science education, and the use of science in everyday life. The curriculum development also promoted a view of science as a part of general education, until Year 11 (age 15), when specialisation into the separate sciences occurred. It had a focus on a science education for future citizens in a scientific society, not just a science education for future scientists. In terms of Maori and science education, this curriculum development resulted in: • • •
• •
discussion papers on Maori and science education as part of the curriculum consultation process (Department of Education, 1986a, 1987–89); meaningful contexts for Maori students learning science (Bell, 1990; Ministry of Education, 1990a; McKinley, McPherson Waiti and Bell, 1992); the writing of science classroom resources in te reo Maori and associated vocabulary development (Department of Education, 1988b; Rikihana, no date; Trinick, no date); a teacher development guide titled Science Aotearoa ( Jesson, 1990; Ministry of Education, 1990d). a policy statement in the new draft science curriculum on the teaching and learning of science in te reo Maori (Ministry of Education, 1990a): M2ori students and science education Science education needs to make science accessible to more Maori students and to reflect the spirit of the Treaty of Waitangi. The Treaty of Waitangi guaranteed the tangata whenua possession of their taonga, of which language, te reo, is one of the most important. Maori people, therefore, have the right, if they wish, to learn science in the context of their world and in te reo Maori. For many students, this means that the science being learned in schools will be linked to experiences in their daily lives, using examples from te ao Maori. For students in bilingual and kaupapa Maori schools, science will be learnt in te reo Maori me ona ahuatanga, as part of an holistic approach to learning.
160
Culture, M1ori and science education
Teachers need, in their teaching and learning activities, to value the use of Maori language and to acknowledge the experiences of Maori students. This acceptance will strengthen the students’ self-esteem by affirming their identity as Maori, and will help them take their part in the world of science. While science education may begin in the world of the Maori students, it also needs to help them go beyond this. Teaching and learning activities and school programmes need to enable all Maori students to gain access to science and reach their full potential in science education (Ministry of Education, 1990a: 13). Although this document was not made an official syllabus due to a change in government and a new, more conservative Minister of Education, it did facilitate some changes with respect to Maori and science education. When this draft syllabus was released for comment in June 1989, the statements on Maori students and science education drew a lot of discussion, including racist comments (Bell, 1990). The debates in the media included the view that there was no place for Maori knowledge in the science curriculum. However, ‘by the time the English language version of the latest curriculum was published in 1993, not only was the addressing of Maori students compulsory for the contractor but the sections on Maori students and science education drew very little response (Haigh, 1995a)’ (McKinley, 1995b: 34). Therefore, the 1993 science curriculum, building on the 1990 draft, was one of the first official curriculum policy documents that has addressed Maori issues and included Maori knowledge, over 150 years after the signing of the Treaty. Bicultural and bilingual science education The national curriculum developments in the 1980s, as discussed in the previous sections, can also be seen as bicultural and bilingual science education (McKinley et al., 1992). This was in contrast to the prevailing view of integration in educational policy at the time. Bicultural science education was seen to include the following: •
Using examples and contexts from the students’ culture One of the first attempts to make learning in science more relevant and interesting for Maori students was to use Maori examples to illustrate the science ideas in the classroom. The examples came from traditional Maori life, for example celestial navigational techniques, and current Maori life, for example cooking food in a hangi (an earth oven). Another way was to use Maori contexts for teaching and learning the science, for example wairoa (water) and kaimoana (sea foods). As McKinley, McPherson Waiti and Bell (1992) state: ‘The goal was to include Maori examples and contexts which would be perceived as more interesting, relevant, and useful by Maori
Culture, M 1ori and science education
•
•
•
•
161
students, without omitting any of the basic content (both concepts and skills) in the national science curriculum’ (p. 583). Using different teaching, learning and assessment activities The use of more culturally appropriate teaching, learning and assessment activities is one direction science teachers have been exploring, in an attempt to improve achievement and take into account the culture of Maori students. Maori culture values cooperative, oral and group-orientated ways of working rather than ways based on individual competition (Hemara, 2000; Pere, 1982). Reviewing the aims of science education The science curriculum review of the 1980s (Bell, 1990) had as a central aim ‘a science for all students’. The interpretation of the aim was for all students, including Maori and girls, to study science in everyday contexts until the age of 15 (when most students take their first national qualification examination or assessment). ‘While Maori parents and educators want a science education that takes into account the cultural backgrounds of the students, they also wish Maori students to have an education that would enable them to compete in the scientific job market in New Zealand and internationally’ (McKinley, McPherson Waiti and Bell, 1992: 585). Using indigenous science in science education The inclusion of Maori science (as indigenous science) in the science curriculum was to give value to Maori cultural knowledge. Science (both western and indigenous science) is seen as a human construct and as culture-dependent. ‘If science is viewed as a human construct then common ground between indigenous science and modern science can be described’ (McKinley, McPherson Waiti and Bell, 1992: 586). The articulation of M4ori cultural knowledge for scientists and teachers of science In being a citizen of the bicultural nation of New Zealand, one has access to both Maori and Pakeha cultural knowledge in making educational decisions. There was an awakening, during the 1980s curriculum developments, by Pakeha to Maori demands as to the importance of considering the cultural knowledge of both Maori and Pakeha, as the two Treaty partners in the founding of the nation. ‘It was a demand for the legitimacy of Maori epistemology . . .’ ( Jesson, 1990: 51).
In 1993, the annual journal of the Waikato group (SAMEpapers) published two articles on the need to take into account both Maori and Pakeha cultural knowledge – in this instance, in ecological management. Haami (1993) and Roberts (1993) each wrote of the kiore rat (Rattus exulans): Haami from a Maori perspective and Roberts from a Pakeha perspective, to highlight the ‘inadequacy of our knowledge base – and hence our decision making – when it is based solely on either one of these two perspectives . . . in emphasising the importance of cultural as well as scientific knowledge, it is intended to highlight the need – and
162
Culture, M1ori and science education
indeed the right – of all New Zealanders to have access to both perspectives’ (Roberts, 1993: 23). The pedagogical knowledge of science teachers in New Zealand could well include both scientific and Maori cultural knowledge. In highlighting the cultural significance of the kiore for Maori, Haami (1993) discussed Maori cultural knowledge and traditions relating to the kiore rat: To the Maori people of Aotearoa New Zealand, the kiore rat (Rattus exulans) was an historically important food source. In a land of no indigenous terrestrial mammalian fauna, the value of this resource extended beyond that of nutritional necessity. Kiore were used as currency in the acquisition of land, and in the naming of people and places. They were sung about in waiata (songs) and haka (dances), and knowledge of their habits was incorporated into whakatauaki (proverbs) as well as being depicted in carvings. These cultural traditions reveal an attitude towards rodents which is in distinct contrast to that of New Zealanders of European descent. (Haami, 1993: 5) Tradition says that the kiore were brought to Aotearoa in the Aotea waka (canoe) as a food source, and for this purpose reserves were set up by each iwi (tribe) where kiore were bred for kai (food). This information reveals that the kiore is remembered in almost all aspects of Maori tikanga (customs) not as a nuisance, or worse, but as a thing of some considerable value . . . Most Maori I spoke to in the course of this research thought the kiore was extinct, and on learning that it still survived (though only on some off shore islands and in the very remote south-west corner of the South island), they shed tears of joy, because it carried their thoughts back to the time of their tupuna . . . To many Maori the eradication of the kiore by rats and other predators introduced by the Europeans is paralleled by the demise of our Maori culture. (Haami, 1993: 19) In contrast, Mere Roberts (1993) wrote of the European view of the rat, outlining the cultural attitudes of people of European descent towards rodents, as carriers of the two diseases of the plague and typhus. The recurrent and catastrophic nature of these diseases had etched themselves deep into the culture and psyche of the Europeans, with a fear of rodents prevailing. No diseases have yet been isolated from R. exulans in New Zealand, and as such it poses little or no threat of disease to humans in New Zealand, since the current distribution of this rat is mainly on offshore islands, most of them uninhabited. However, the impact of Rattus exulans on certain New Zealand indigenous flora and fauna has led to it being labelled as a pest and targeted for extinction. The New Zealand Department of Conservation has been involved in the eradication of this species since 1978. As it previously had not had protected
Culture, M 1ori and science education
163
status, no management plan for this species has been developed. Calls are being made for such a management plan to be developed, based on both scientific knowledge and the cultural knowledge of the indigenous people, particularly when it is regarded by them to be of historical significance (Roberts, 1993). In summary, such a bicultural approach to science education has been criticised in New Zealand ‘as not helping the achievement of Maori students, and in fact may serve best to raise the cultural awareness and sensitivity of non-Maori. All Maori are bicultural . . . Maori view bicultural and anti-racist science education as essentially a problem for non-Maori to address’ (McKinley, McPherson Waiti and Bell, 1992: 587–8). In the 1980s, for Maori, the top priority was the survival of te reo Maori and increasing the self-esteem and achievement of Maori students through the use of their own language as the language of instruction and learning in science education. The use of the Maori language did not equate to a direct translation of the scientific words in English. For example, Maori words often carry the context with them, unlike in English where the context is derived from the surrounding words. There are 150 words in Maori for eel, depending on where it is found and what time of the year it is (McKinley, McPherson Waiti and Bell, 1992). The goal was not to produce monolingual students, speaking Maori only. English was also taught as a second language. The dominance of English as the language in most media (newspapers, magazines, radio, television, computer games) in New Zealand also ensures that Maori educated in total immersion and kura kaupapa schools are bilingual.
164 Culture,8M1ori and science education: research and development Chapter
Culture, M2ori and science education Research and development
National curriculum developments in the 1990s In the (current) New Zealand 1993 science curriculum, written in English, the statement on Maori and science is as follows: M2ori and Science Science education needs to make science more accessible to Maori students. It must make use of teaching strategies which are effective with Maori students and must be responsive to the diversity of their cultural and language backgrounds. Acknowledging tikanga Maori, and valuing the use of Maori language and the experiences of Maori students, affirms their identity and creates a positive learning environment. An inclusive curriculum in science provides opportunities for Maori students to: • • • • •
• • •
learn science that they, their peers, their teachers, their whanau, and the wider community value; learn science through the medium of te reo Maori; learn science which acknowledges and values Maori scientific knowledge; develop scientific concepts within Maori contexts; use their preferred learning and communication styles, such as cooperative learning and holistic approaches; and have oral contributions recognised for both learning and assessment purposes; interact in an environment where the language and resource materials used are non-racist; use a wide range of resources in te reo Maori; have access to positive Maori role models, including Maori teachers, in their science programme. (Ministry of Education, 1993b: 12)
In 1992, the landmark decision was taken to write national curriculum documents in Maori. In 1996, the Ministry of Education published a science curriculum (Te Tahuhu o te Matauranga, 1996) in te reo Maori. Previously, not
Culture, M1ori and science education: research and development
165
even the Maori language curriulum had been written in Maori (McKinley, 1995b). The curriculum was developed under contract to the University of Waikato, with Pauline McPherson Waiti and Liz McKinley as the codirectors, in collaboration with a team of writers, a kaumatua, support group, consultations groups, Ministry of Education staff, Te Taura Whiri i te reo Maori (the Maori Language Commission), teachers and students. The curriculum document was developed as a response to Maori concerns over the failure of the current education system to meet their aims and aspirations, and the low achievement by Maori in national examinations. In her masters thesis, Liz McKinley critiqued this development of Te TauAki Marautanga PUtaiao: He Tauira – the draft version of Science in the New Zealand Curriculum: Curriculum Statement in Maori. Her thesis contained three main arguments: First, I argue that Maori have had an influence on the emergence and execution of curriculum policy written in Maori. The second argument is that the discourse of scientism is still heavily implicated in the construction of this policy and that Maori knowledge has not been legitimated even though Maori language has been used. This can be interpreted as a cooption of Maori language for regulation and control. Thirdly, I argue that the benefits for Maori as a group to come out of this development were in the form of technical language development, teacher development and as a resource for teachers. (McKinley, 1995b: ii) The thesis argued that the draft science curriculum policy Te TauAki Marautanga PUtaiao: He Tauira represented the intersection of three major discourses: a discourse associated with ‘new right’ policies that had been implemented in education; Maori views of education; and a scientific view of curriculum construction. Because the policy documents arose from all three discourses and did not arise out of a Maori discourse only, the document did not address some important curriculum issues for Maori ( McKinley, 1995b). She argued that the relationship between Pakeha and Maori, of power and domination, is evident in different aspects of the curriculum and curriculum development process. First, it was noted that the level of negotiation in the development process was such that it cast doubt on whether this project could be called a curriculum development project at all. The State did not negotiate with Maori on the content of the curriculum. The task was perceived as one of translation, not the development of a science curriculum that included, validated and legitimated Maori scientific/cultural knowledge as well as western scientific knowledge. McKinley (1995b) felt that Maori discourses had a very low input with respect to the structure of the curriculum, but a much larger input on aspects that the government wanted little or no stake in, for example the language development or how the consultation would be carried out.
166
Culture, M1ori and science education: research and development
The relationship between Pakeha and Maori of power and domination was seen as inherent in government institutional structures, for example, a national curriculum. For example, the development of the science curriculum in Maori came sixteen months after that for the English version. The same structure of levels and strands and the use of learning outcomes was imposed on both the English and Maori versions. However, the development of the Maori version had other aspects imposed on it as the developers were directed to maintain the achievement objectives (Ministry of Education, 1993b) already constructed in the English version. ‘The restrictions on the development of the Maori version meant that the Maori group could not develop a curriculum from the same starting point that the group writing the English version did. Hence, in the Maori version, Maori did not get to enter negotiations at the same level as those involved with the English version’ (McKinley, 1995b: 54). Their needs were considered by the Ministry of Education to have been met by the English version and Maori were not given the same negotiating power as the English version writing team. The curriculum development process was seen as a translation. ‘This implies that Maori were not even one of the major stakeholders in their own curriculum development’ (McKinley, 1995b: 54). Hence, it is not considered to be a Maori curriculum. The relationship between Pakeha and Maori of power and domination extended to Pakeha knowledge being included in the curriculum, but not Maori cultural knowledge. The Maori writers had to support the assumptions about the nature of science, the nature of the world and the nature of science education found in the English version of the document and this was seen as a ‘further attempt to incorporate Maori into the dominant discourses’ (McKinley, 1995b: 60). This relationship between Pakeha and Maori was also evident in the process of curriculum development as the developers had to accept a linear model of curriculum development, with the different learning areas being developed one after the other and not together for holistic development. This impacted on the ideas on the nature of science and epistemology which were to underlie the curriculum and which were ‘incompatible with Maori epistemology. The constraints were such that only “Maori knowledge” that fitted into the overall paradigm could be included and anything that Maori may have wanted to include that did not fit the dominant discourse of science had to be treated as an extra. Such measures lead only to superficial changes in the curriculum for Maori and do not challenge the epistemological core of “western scientific rationality” ’ ( McKinley, 1995b: 60). For example, in the English version, one of the strands is called ‘Making Sense of the Living World’. The use of the word ‘living’ defines what the living world is and the knowledge in this strand is the basis for the understanding of it. In comparison, the equivalent strand in the Maori version is called o Mataroa which is translated as ‘from the living world’ . . . This is done with the recognition that what is contained in this strand does not in any way define, by itself, what
Culture, M1ori and science education: research and development
167
the living world is but the knowledge contained in the strand contributes to the overall knowledge of the living world. In addition, the 5 Mataroa strand has included aspects of earth science, such as forces that shape the Earth (Papatuanuku) that are not classified as ‘living’ under scientific classification. (McKinley, 1995b: 63–4) Liz McKinley discussed the different views of the nature of science as yet again indicative of the relationship between Pakeha and Maori as one of power and domination. In contrast to the realist, objectivist, positivist view of science, a Maori world view seeks to explain the origin of the universe through whakapapa (genealogy) and the personification of natural phenomena (Pere, 1982). ‘Maori developed complex genealogical constructs to explain and link the time before (Te Kore), during (Te Po) and after (Te Ao Marama) the origin of the universe . . . The universe is seen as holistic and dynamic; there is within it an ongoing process of continuous creation and re-creation’ (McKinley, 1995b: 70). A second important idea in the Maori world view is that there is ‘no demarcation between the supernatural and the natural; both are part of the unified whole, (Pere, 1982). Hence, in direct contrast to traditional western science, indigenous knowledge is thought by many scientists, science teachers and policy makers alike to contain no science beyond a simplified localised empirical gathering of information for the direct purpose of survival in the immediate environment (for example see Harding, 1993). For example, maramataka (fishing and planting calendars) are often seen by scientists and science teachers in this light’ (McKinley, 1995b: 70). In addition, the Putaiao document did not address the general question of what qualifies as scientific knowledge (Barker, 1999; McKinley, 1995). It was not in the brief given to the contract holders, although it did arise in the debates of the writing group. Therefore, the document ‘leaves unchallenged what Harding (1993) claims is the common assumption, viz. that the knowledge of indigenous people is closed, pragmatic, utilitarian, value-laden, indexical and context-dependent’ (Barker, 1999: 56). The Pakeha and Maori power relationship is also evident in the psychological basis to the curriculum in which Maori students have been pathologised (become the abnormal) (McKinley, 1995b). The curriculum is underpinned by discourses of teaching and learning, including the traditions of developmental psychology, behavioural psychology and constructivism, each with hidden assumptions about people. The discourse of Maori education had had no effect on the teaching and learning theories implied in this document. As Valerie Walkerdine (1984: 154) stated, ‘psychology’s claims to truth are premised on the constitution of the individual as an object of science in certain historically specific conditions of possibility’. Hence, as McKinley (1995b) points out, ‘not only does the child become pathologised, but any close connections with a culture that is not inherent within the theoretical framework becomes a deviation from the norm . . . Hence this way the developmental psychology of the 1960s and which is the basis for curriculum development in the 1990s, will always find
168
Culture, M1ori and science education: research and development
Maori in deficit. There is no other interpretation able to be made within this type of framework’ (McKinley, 1995b: 84). Maori students were seen as objects, not subjects, of science (McKinley, 2003). In summary, McKinley stated that while Maori knowledge may be incorporated into the curriculum, Maori knowledge (other than what science defines as being suitable to science content) has not necessarily been legitimated. The connection between the language and knowledge is implicit – the assumption is that the learning of Maori language will bring with it Maori knowledge. . . . The document as a whole can be seen as a contradiction. On one hand some Maori issues are seemingly being addressed, such as the use of Maori language in curriculum policy documents. However, on the other hand there are some very dubious constructs with respect to Maori being used in the psychologically based teaching/learning theories . . . Many Maori are still a long way from gaining the sort of autonomy that they wish to have in Maori education. (McKinley, 1995b: 85) The development of Maori vocabulary was a major outcome for Maori of the 1990s curriculum development, with the Maori Language Commission being involved. It was ‘an opportunity for systematic development of science vocabulary – something that had not happened before’ (McKinley, 1995b: 57). In 2000, the Ministry of Education published a book in Maori to help teachers of science use the science curriculum document, Putaiao (Te Tahuhu O te Matauranga, 2000).
Research for M2ori science education In 1989, the restructuring of education administration in New Zealand saw the disbanding of the Curriculum Development Division, and I returned to the University of Waikato to become the first full-time staff member of the newly established Science and Mathematics Education Research Centre (now the Science and Technology Education Research Centre). Within two years, both Elizabeth McKinley and Pauline McPherson Waiti joined the staff of the University and we were able to continue discussions on Maori science education with colleagues. At this time, it was noted that the Learning in Science Projects, including the work by Stead (1984), had not researched culture as a major variable in studying children’s understanding of scientific concepts (McKinley, 2000). Instead, ‘science education has largely confined cultural issues to matters such as “science and society” and “science for all”. Only recently have educators begun to argue that science itself is a culture whose knowledge, values, and processes create problems in today’s multi-cultural schools’ (McKinley, 2000: 34). From 1990, the research at Waikato began to explore the debates and issues surrounding culture and science education in Aotearoa New Zealand. At the
Culture, M1ori and science education: research and development
169
time, there was a strong view that Maori educational research needed to be done by or directed by Maori, so that the research was for the benefit of Maori and not as the ‘researched’ (McKinley, McPherson Waiti and Bell, 1992). This was in response to Maori being one of the most researched people in the world, with the researchers benefiting from the research, rather than those who have been the object of the study. The non-Maori researchers had questions of their own, which resulted in the creation of knowledge about Maori that is radically different from how Maori wish to define or construct themselves (Bishop and Glynn, 1999). It was felt that Maori needed to have a say on who should be doing the research, how it should be done, what should be researched, the validation of findings and the use of the findings (McKinley, 1992a). In the 1990s, research into Maori science education at the Centre was done by Maori and Pakeha, but it was addressing the concerns of Maori, not Pakeha. Science education research by Maori is scarce as few Maori are trained as both science teachers and researchers and also fluent in Maori. There has been a call for more Pakeha to be involved, Pakeha who are knowing of Maori knowledge, language and customs (McKinley, 1998). Most Maori students learn science in English in mainstream schools and are taught by Pakeha. Pakeha science teachers of Maori students also wished to explore in their masters theses new teaching and assessment strategies that meet the aspirations, preferences and practices of Maori, as a way to addressing the low achievement of Maori students in their classes. Examples of such research are Gribble (1993) and Kent (1996). The views and work of the Waikato group on Maori research in science education in the early 1990s was documented in McKinley, McPherson Waiti and Bell (1992). The paper proposed ‘that an area of future research in science education is that of learning science in an indigenous language’ (McKinley, McPherson Waiti and Bell, 1992: 579). The following areas were thought to be worthy of further thought and research: •
• • • • • •
In what ways and to what extent are the learning outcomes (cognitive and affective) in science influenced by the learning of western science in te reo Maori? What aspects of traditional Maori science are appropriate for inclusion in the primary and secondary curriculum? What is the modern equivalent of traditional Maori science? To what extent and in what ways has intellectual colonisation of Maori in science occurred? How can the school curriculum best be organised and what resources are needed to enable science to be learnt in te reo Maori? What is the role of Maori contexts in the learning of science? What is the effect of students learning both Maori science and western science in the science curriculum? What would a both-ways education be like? Would it mean learning western science in English and Maori science in Maori? Is it more important that we
170
•
• •
Culture, M1ori and science education: research and development
are able to explain concepts of science in Maori rather than be able to translate all technical scientific terms, for example, photosynthesis? For the students, will the two kinds of science knowledge be distinct, overlap or be integrated with each other? To what extent would it be helpful for Maori students to learn about western science in the Maori language? What are the issues in vocabulary development for teaching and learning science in Maori as perceived by Maori elders and parents, Maori teachers of science, Maori science students and Te Taura Whiri i te Reo Maori (the Maori language Commission)? What teaching, learning and assessment activities best help Maori students learn science? In what ways and to what extent does bicultural and anti-racist science education influence the learning outcomes (cognitive and affective) in science education for non-Maori students? (McKinley, McPherson Waiti and Bell, 1992: 592)
These ideas were further debated at an international conference entitled ‘Language, Culture and Science Education’, which was held at the University of Waikato, in July, 1992, with some of the key papers being published in a special issue of the Centre’s annual journal, SAMEpapers 93 (McNaught, 1993; Moli, 1993a; Rutherford, 1993). The purpose for the conference was twofold: one was to highlight the research, development and discussions ‘important to indigenous peoples by indigenous peoples in the fields of science and mathematics education’ (McKinley and McPherson Waiti, 1993: 1). The second purpose was to provide a forum as support for the research beginning in the Centre on Maori and science education, for example the masters study and research by Liz McKinley and Pauline Waiti. The main debates were around the issues of highlighting concerns for Maori about the underachievement of Maori students in science, for example ‘the use of Maori language (and other indigenous languages) as the medium for instruction in educational institutions’ (McKinley and McPherson Waiti, 1993: 1); the use of a Maori knowledge base in the science curriculum; science vocabulary in Maori; the nature of science; and teacher education (McKinley and McPherson Waiti, 1993). The remaining part of this section reviews the other research done in the centre in the 1990s. Leeana Kent (1996) (a Pakeha) explored the issue of the nature of an assessment task which takes into account Maori cultural values. What are the preferred assessment methods for Maori? What assessment practices and their purposes reflect kaupapa Maori values? It was thought that present assessment methods are not reflecting the scientific cognition of Maori students. At the time that the Maori science curriculum was published, the assessment and curriculum policy documents from the Ministry of Education (for example Ministry of Education, 1993a, 1993b, 1994b) included statements about using assessment practices that were fair to all students and which recognise the differences in gender, culture, background and experiences that students bring to their learning.
Culture, M1ori and science education: research and development
171
The aim of the research was to explore an effective assessment strategy to ascertain Maori student understanding of the phenomenon of heat. The thesis proposed that, for Maori students, one form of assessment (the individual interview) would be a more successful and valid means of assessing the knowledge and understanding of the scientific concept (heat) than a pen and paper task. As Kent stated: Interviews can be seen as a legitimate form of assessment for Maori students. Maori have a strong oral history. In the past ten years, there has been a move by Maori to include oral assessment techniques in their evaluation of Maori student learning. For example, assessment of Maori students for School Certificate te reo Maori includes the use of interviews between the students, their classroom teachers and a ‘member of a regional team of teacher connoisseurs’ . . . a holistic approach to assessment is taken, but the format is Pakeha. The interview allows the student to use culturally appropriate verbal and non-verbal communication with the interviewer, in an atmosphere which effectively reduces candidate-stress and apprehension. . . . The processes of listening and observing that are a major part of interviewing are important within kaupapa Maori. As most science teachers are Pakeha, there are cultural barriers between the majority of Maori science students and their teachers. Cross cultural assessment can best be carried out by the teacher within the teaching–learning process . . . but to do so also requires an empathy with the student, and an acceptance, if not an understanding, of the different knowledge systems. Interviews encourage teachers to listen to the student responses with respect, and encourage students to communicate positively with the teacher. Together they can increase the student–teacher rapport in a bicultural situation, as between Maori and Pakeha. (Kent, 1996: 203–6) The research involved a Year 9/10 (age 13–14 years) bilingual whanau (family-like) science class (n=7), from a co-educational secondary school and the teaching focused on the topic of heat and the hangi (earthen oven). The conceptual learning of the students during the unit of work was assessed by a summative assessment task, which was a student and researcher collaboratively designed, written, open-ended response, achievement-based assessment task. The researcher also assessed the use of an interview based on their written responses, together with further probing for understanding. The written response data and the interview data were analysed for differences in the student responses between the written assessment and the interview. The data analysis strongly suggested that the interviews were a more valid method of assessment compared with the written assessment task for these Maori students. The interviews were able to elicit a greater depth of scientific understanding among the students than was apparent in the written responses. The
172
Culture, M1ori and science education: research and development
students’ difficulties with some of the skills required for written assessment, for example the conceptualisation of the task and the structuring of the responses, were able to be surmounted with the use of rephrased questions, analogies and probe questions in the interview. All the students agreed that the individual interview was a better method to determine their understanding of the phenomenon of heat, as it helped them understand the assessment task better, and as they had difficulty expressing themselves in writing. The interview could also be used as a teaching tool, when used for formative purposes, as it allowed ‘the researcher to introduce otherwise unused scientific terms, reveal student alternative frameworks, and develop the presence of analogies to enhance learning through scaffolding and bridging between the alternative frameworks and the scientific conceptions’ (Kent, 1996: 200). Wayne Gribble (1993), a Pakeha, used the contexts of Maori myths and legends during the teaching of a unit on volcanoes and earthquakes to two Year 9 classes, based on the ideas of McCraw (1990). The study aimed to explore a possible means of increasing the cognitive learning made by Maori students in science by introducing Maori legends into science classes, thereby taking into account Maori students’ existing ideas, experiences and cultural beliefs, as a starting point into science. The cultural knowledge of Maori was not to be replaced by the newly learnt scientific knowledge as suggested by Jesson (1990) but to stand alongside it and provide a link between Maori cultural knowledge and western scientific knowledge. The study also explored the effect of using Maori legends on the non-Maori students in the classes. Two year 9 classes (aged 13–14) from a co-educational secondary school were involved in the study. Each class had four science lessons, taken by a Pakeha teacher. In one class (n=30) the teaching combined Maori legends with an earth science unit, while the other class (n=20) was taught the normal scheme. Pre- and post-surveys were completed by all students to assess their attitudes towards Maori myths and legends and the topics related to earth science, as well as to assess their cognitive learning. Pre- and post-interviews were also carried out on a sample of students from each class to provide further data for the research. The three-week teaching unit was on volcanoes and earthquakes. Two Maori legends were used, one in the first lesson, ‘The Warrior Mountains’ by Katerina Mataira (1982), and one in the third lesson, the legend of Ruamoko (a Maori god of volcanoes and earthquakes). The findings indicated that the Maori students from the intervention class made no significant improvement in cognitive learning from the teaching of the earth science unit, which included the two Maori legends, than students from the comparison class. A similar trend was found for the non-Maori students. Students from both classes held strong views about the nature of science and what should be included in science, with many students initially holding the belief that Maori legends did not have a place in the science classroom. In the post-survey findings, the Maori students in the intervention class had a positive attitude towards the inclusion of aspects of Maori culture in science
Culture, M1ori and science education: research and development
173
lessons and the value that was placed on their cultural knowledge by a Pakeha teacher. This improved the rapport between the Maori students and the teacher (who was also the researcher). In the post-survey findings, a significant number of non-Maori students in the intervention class changed their views to believing that myths and legends did have a place in the science classroom. Given the small sample size and the short span of the teaching, any test of significance must be interpreted cautiously. These findings parallel studies on Taha Maori (see Bishop and Glynn, 1999), which indicate that while non-Maori gain knowledge and acceptance of Maori cultural knowledge, the achievement of Maori is not increased. Pauline McPherson Waiti researched the role of support systems in addressing Maori underachievement and non-retention in science education. She researched learning outcomes of a wananga or science camp held by the National Association of Maori Mathematicians, Scientists and Technologists (NAMMSAT), in association with Te Puni Kokiri (the Ministry of Maori Development). The camp was for 34 Year 8 (aged 12) kura kaupapa Maori students and their teachers in a Maori context (Waiti, 1994). ‘This Maori context not only involved the immediate surroundings of the marae but also the majority of the facilitators were Maori, the language of the wananga was Maori and aspects of the Maori scientific body of knowledge were included in the programme’ (Waiti, 1994: 7). The aims of the wananga were to provide professional development for the teachers and to provide some science resources that could be used in the kura. For NAMMSAT, the main purpose was to foster the students’ interest and participation in science. The students were given pre- and post-interviews in te reo Maori and they kept a journal during the camp. The findings suggested that the students broadened their views about scientists and the work they do during the time of the camp. More importantly, they could now include themselves in the picture of scientists doing scientists’ work (Waiti, 1994). In 1992, Liz McKinley (1992b) researched the concerns that eight teachers had about teaching science in Maori in total immersion schools, so as better to inform teacher development and pre-service programmes for teachers teaching science in Maori. The areas that the teachers felt needed consideration when developing courses for practising and student teachers were the Maori knowledge, the science content, the Maori language for the science vocabulary, the pedagogies used, the Maori language fluency of the tutor and confidence to teach science.
Ongoing research The research at Waikato continues and this ongoing research is detailed in this section. This research can be loosely grouped into two main situations: •
research in kura kaupapa and total immersion schools, where putaiao (science) is taught and learnt in te reo Maori;
174
•
Culture, M1ori and science education: research and development
research in mainstream schools, where Maori science students are taught in English and usually by a Pakeha teacher.
The research is informed largely by two different views of ‘post-colonial’. 1
2
A post-colonial view as the one developed after colonisation by Pakeha in 1840, with the signing of the Treaty of Waitangi. The impact of colonisation on Maori was mentioned by the Waikato group as early as 1992, when it was acknowledged that, like the colonisation of other indigenous peoples, colonisation of Aotearoa included the loss of land, the loss of language, and the consequent loss of culture in the home and school situation, the destruction of traditional social order, including informal education structures in the extended family (McKinley, McPherson Waiti and Bell, 1992). A post-colonial view of Maori and science education is to view Maori students as taking up many different (multiple hybrid) positionings in the discourses (McKinley, 1992a). Using the term ‘post-colonial in this second sense, Liz McKinley and Nesta Devine (2001) outlined ways in which a post-colonial perspective on Maori and education can present different subjectivities for Maori students. They introduced ways to disrupt and problematise a singular and often stereotypical idea of what it is to be Maori in the (science) classroom, including: • • • •
Maori students’ experiences of the classroom as distinct from teachers’ experiences of Maori students; high-achieving Maori students being seen as Maori and not white; being Maori in mainstream education situations is different from being Maori in kura kaupapa situations; ‘Maori’ may not exist in the context of kura kaupapa as there is no need to distinguish ‘Maori’ from others. ‘Maori becomes the unspoken, assumed, norm and it may be that, arguably, tribal structures become more prominent. The national curriculum documents indicate this where statements concerning, for example, Maori and science education in the English version have been omitted in the Maori version’. (McKinley and Devine, 2001: 5)
They asked questions to disrupt and problematise: • • • •
In what ways do Maori students make meaning of their experiences in the science classroom? What Maori children’s voices on their experiences in the classroom are known to science teachers? What are parents’ voices on their Maori children’s experiences in classrooms or schools? Do Pakeha mothers of Maori children and Maori mothers of Maori children have different views of their children’s experiences in the classroom?
Culture, M1ori and science education: research and development
• •
• •
175
What do Maori students construct as ‘achievement’, given that achievement is a culturally constructed concept? How do teachers make sense of and explain what is going on in their classrooms, especially, for example, when there is a tension between recognising the individual, as in western theorising, and recognising the group, as in Maori theorising? Are there differences between the way Maori and Pakeha teachers see the experiences of Maori children? How do the different discourses used by teachers influence their meaning-making of the classroom experiences of Maori students, for example Te Ao Maori, learning theories and styles, behaviour modification, pragmatism, deficit theory, the ‘New Right’ reforms?
At this same time, international debates on cultural and ethnic diversity in the science classroom have included the notion of cross-cultural science teaching and border crossing (Aikenhead, 1996). The notion of border crossing ‘recognises the inherent border crossing between students’ life world subcultures and the subculture of science, and that we need to develop curriculum in mind, before the science curriculum can be accessible to most students’ (Aikenhead, 1996: 2). An effective teacher is positioned as a culture broker (Stanley and Brickhouse, 2001). A critique of these ideas, arising from an analysis of the New Zealand situation, is that there is a problem in viewing any encounter simply being ‘managed’ as pedagogical moments requiring racial, cultural or gender sensitivity. In this model of cultural diversity, communication between the participants, teachers and student, is simply technical. Furthermore, it is thought that misunderstandings arise because the parties are culturally, racially, physically, mentally or sexually different. All science educators need to do is to learn how to work their way through these differences. (McKinley, 2001: 74) But this viewpoint does not address racism, the responses of subordinate groups, the effects of colonialism or the power relationships of dominant and subordinate groups marked by histories of oppression. In this view, the notion of cultural diversity is an epistemological object and underlying it is the assumption that ‘we can stand outside the hierarchical social relations that exist in our lives and, as innocent subjects, we are not accountable for the past or implicated in the present (systems of oppression)’ (McKinley, 2001: 75). A disruption of the hegemonic ways of seeing are needed. Other research at the University of Waikato is by educational researchers Russell Bishop and Ted Glynn (1999). Their thesis is that the power relationship of dominance and subordination between Pakeha and Maori in the classroom needs to be changed if Maori achievement is to be raised. They particularly
176
Culture, M1ori and science education: research and development
focus on addressing the power imbalance resulting from the use of both Maori and Pakeha cultural knowledge and sense making (ways of knowing) in the classroom. Obviously, the professional decision making of teachers is crucial in deciding how ‘both’ will be manifested. They propose the use of metaphors to establish new power-sharing relationships, for example between teacher and student. They themselves use the Treaty of Waitangi as a metaphor as well as a model for establishing power-sharing relationships. The metaphor has three aspects: • • •
partnership: Maori people guaranteed a share in the power of decision making; protection: Maori people guaranteed the power to define and protect knowledge and language; participation: Maori people guaranteed benefits of participation in education. (Bishop and Glynn, 1999: 199)
They refer to the work of Graham Smith (1997) who in his research on kura kaupapa Maori introduced six new metaphors for educational relationships and interaction patterns, as discussed previously: 1 2 3 4 5 6
Tino rangatiratanga (relative autonomy/self-determination); taonga tuku iho (cultural aspirations); ako (reciprocal learning); kia piki ake i nga raruraru o te Kainga (mediation of socio-economic and home difficulties); whanau (extended family); kaupapa (collective vision, philosophy).
Here, the term ‘metaphor’ is not used as synonymous with ‘analogy’, but as the way ‘we organise our relationships . . . and how we, and those with whom we interact, understand or ascribe meaning to particular experiences and what eventually happens in practice’ (Bishop and Glynn, 1999: 166–7). They argue that these new metaphors are needed to create new ways of relating between teachers and students in the classroom. They advocate the use of metaphors created in kaupapa Maori educational contexts: kohanga reo, kura kaupapa Maori and Kaupapa Maori research. Examples have included whakawhanaungatanga (the process of establishing relationships in a Maori context) as a metaphor for creating family-type relationships, and hui (meeting) as a metaphor for collaborative storytelling, leading to narrative pedagogies (Bishop and Glynn, 1999). These practices set up learning–teaching relationships where: •
culture counts: classrooms are places where learners can bring ‘who they are’ to the learning interactions in complete safety, and their knowledges (including languages and language patterns) are ‘acceptable’ and ‘legitimate’;
Culture, M1ori and science education: research and development
• • • • • •
• • •
• • • • • • •
177
learners can initiate interactions; learners’ rights to self-determination over learning styles and sense-making processes are regarded as fundamental to power-sharing relationships; learners are able to be co-inquirers, that is, raisers of questions and evacuators of questions and answers; learning is active, problem-based, integrated and holistic; learning positionings are reciprocal (ako) and knowledge is co-created; classrooms are places where young people’s sense-making processes, languages and knowledges are validated and developed in collaboration with others; teachers and learners interact and exchange roles; assessment practices employ a wide range of culturally appropriate and culturally constructed strategies; metaphors of conversation, particularly those that incorporate ‘not knowing’ and ‘collaborative storying’ guide the development of principles and practice, rather than metaphors of transmission; motivation is intrinsic to the collaborative achievement of tasks and to the co-construction of meaning; critical reflection is part of an ongoing critique of power relationships; understandings are related to the experiences of all learners and learners can be aided to become independent, through processes of scaffolding; understandings are gained in real-life (or close to) situations; students are introduced to the variety of discursive processes that create knowledge/s-in-action; problem solving, critical thinking and creative analysis are seen as lifelong skills; teachers are inextricably connected to their students and the community school and home/parental aspirations are complementary. The community and home validate and support the academic success of students. (Bishop and Glynn, 1999: 163–5)
Until now, there has been no research on the achievement of Maori students in science in kura kaupapa Maori. Liz McKinley (2000) listed the research questions before us as we went into the twenty-first century, including the following: •
What makes teaching and learning effective for Maori children? This research should investigate: – – – –
learning and cognitive styles; effective teaching practices and classroom programmes; social and cultural factors (including role models and mentors) that enhance or inhibit motivation and learning; the circumstances that produce (and recognise) exceptional and gifted Maori children;
178
Culture, M1ori and science education: research and development
– – •
What happens if science is taught in te reo Maori? –
•
effective interactions, communications, and relationships between Maori children and science teachers; how Maori children approach learning science.
What are the cognitive effects of teaching science in te reo Maori?
Teacher education What teacher education models are most effective for producing bicultural and bilingual teachers of science? (McKinley, 2000)
By 2003, there were three pieces of research known to the author that were addressing these questions. First, the National Education Monitoring Project (NEMP) had published a report (Flockton and Crooks, 2000) on learning science in Maori. Two-thirds of the 1999 NEMP assessment tasks in science (as well as other subjects) were translated into Maori and administered in Maori immersion settings. The report documented the responses to the science assessment tasks given to Year 8 Maori students in English, who were part of the main sample, and to Maori students educated in Maori in Maori immersion schools or classes within mainstream schools. The assessment tasks addressed the four main content strands of the science curriculum (the living world, physical world, material world, and planet earth and beyond). In summarising the findings, the researchers stated that two of the 18 tasks were found to have posed particular challenges in the Maori language assessments that were not present in the English language assessments due to the translation for the Maori language version. On 12 (75 per cent) of the remaining tasks, Maori immersion and Maori general education students performed similarly. On one task (6 per cent), Maori immersion students performed better than Maori general education students, and on three tasks (19 per cent) Maori immersion students performed worse than Maori general education students. But the researchers added that these comparisons must be viewed with considerable caution (Flockton and Crooks, 2000). At the same time, Georgina Stewart (Stewart, in progress) was researching the ‘effectiveness’ of the Putaiao document (the science curriculum in Maori), since it was published, in raising Maori achievement in science; investigating how children’s learning in science might differ when they learn Putaiao (science) based on a Maori world view; and moving the debates from an understanding of Maori science as either traditional Maori knowledge or as a translated modern/ western science, towards an articulation of a philosophy of Maori science. As kura kaupapa initiatives by Maori are seen as a long-term solution, research is also needed on science education for Maori students in mainstream classes. Whilst the focus on the Maori language as a way of addressing the impact of colonisation gave it a centrality in Maori science educational debates in the two decades of the 1980s and 1990s, research work in the early 2000s is
Culture, M1ori and science education: research and development
179
also focusing on ways of improving the achievement of the majority of Maori students, who do not speak Maori and who are taught in mainstream schools and classes by Pakeha teachers of science. In this work (Bishop, Tiakiwai and Richardson, 2003) researchers are working with Pakeha and Maori teachers in mainstream secondary schools to develop teacher–student relationships that promote Maori achievement. The relationship between teachers and students is seen as crucial in addressing any racist notions of assimilation and deficit theorising, still held by many teachers. Researching the learning of Maori students in mainstream education is also part of a research project begun in 2003. The research project, ‘Maori knowledge, language and participation in mathematics and science education’, directed by Liz McKinley, has a main focus to investigate and propose new ways of developing science and mathematics education that are inclusive of Maori students. The initial work has highlighted the voice of the students and parents/ family in checking how students perceive their own participation, finding out about parents’ and family attitudes towards science and mathematics, and how they are able to support these subject and career choices for their children. In conclusion, the research on Maori and science education at Waikato has, like Maori education in general, been influenced by colonisation notions of cultural superiority and inferiority; assimilationist, integration, multicultural and bicultural education policies of the government; our own stories and the initiatives of Maori to have their needs and aspirations addressed. Hence, the main expression of the international debates on culture and science education in New Zealand has been within the discourses of Maori underachievement and power imbalances. The urgency to address this underachievement remains a government priority, and priority with Maori, who aspire to the partnership of the Treaty of Waitangi being honoured by non-Maori as well as Maori (Bishop, 2000; Glynn, 1998). Mason Durie, in a 2001 speech summarising Maori educational aspirations, stated that educational goals should give support to Maori to live as Maori, be citizens of the world, and enjoy a high standard of living (Durie, 2001). To date, the main debates on culture and science education in New Zealand have been bicultural. There is still the task of researching multicultural approaches to science education. The challenge to the Waikato and other New Zealand researchers and developers is the ways in which the bicultural approach to ethnic diversity in New Zealand classrooms can inform the multicultural approaches also being developed as New Zealand society becomes increasingly multicultural with the contributions of recent immigrants and refugees. However, it is hoped that Aotearoa New Zealand continues to develop its own solutions to the ethnic diversity in our classrooms.
180 Science 9 curriculum development and teacher development Chapter
Science curriculum development and teacher development
Introduction Education may be viewed as a political enterprise (Apple, 1982, 1996; Ball, 1990) and no more so than in curriculum development (Pinar, Reynolds, Slattery and Taubman, 1995), especially in the development of a national science curriculum: Curriculum development is inherently a political process. In it, contestation and negotiation constantly occur with respect to, for example, what content to include; which groups are singled out to be addressed in the document (indeed, the act of singling out groups is political in itself ); how the document should be written, that is, the implied teaching and learning models; which groups/people get an input; and at what level of development that input occurs. The contestation and negotiation between various groups or people can be very fierce. Not just associated with groups that are more politically overt in their aspirations, this political nature is common to the production of all curriculum documents. (McKinley and Waiti, 1995: 75) While the development of school and classroom curricula were often portrayed as a linear, rational, procedural process (for example Taba, 1962; Tyler, 1949), based on teachers’ rational decision making (for example Doll, 1989; McGee, 1997), the development of curricula at the national level may also be viewed as a political endeavour in which the various stakeholders or shareholders contest and negotiate a compromise that is the curriculum. The term ‘curriculum’ refers not just to official documents (the official curriculum) which in New Zealand is a national curriculum. The term ‘curriculum’ also refers to the curriculum developed at the school level (the school curriculum); the curriculum that a teacher intends to teach (the teachers’ intended curriculum or planned curriculum); the curriculum actually taught by the teacher (the teachers’ taught curriculum) (Freyberg and Osborne, 1981); the curriculum experienced by students (the students’ experienced curriculum); the curriculum
Science curriculum development and teacher development
181
as learnt by students (the learnt curriculum); the curriculum actually assessed (the assessed curriculum) and the implicit, unofficial and unintended learning which may occur (the hidden curriculum; Portelli, 1993). Hence the curriculum may be seen as having multiple layers and the curriculum experienced and lived by students may differ from the official curriculum or the teachers’ intended one. In this chapter, the term ‘curriculum’ is used to refer to the understandings and meanings constructed by the teachers and by students in the classroom, that is, the teachers’ and students’ perceived, constructed, lived or experienced curriculum (Bell, 1991b). It is the meanings that students make of the planned learning activities in the classroom that are most important for educators to consider. An account of a curriculum would document not only the content to be learnt (Bell, 1991a) but also the curriculum as a sequence of teaching and learning tasks (Driver and Oldham, 1986); contexts for learning (science) (Bell, 1991a); and thinking skills for conceptual development (Baird, 1984; Baird and Mitchell, 1986). While the official, national science curriculum in New Zealand is changed every 10 years or so, the planned, taught, experienced, learned, assessed and hidden curricula are in constant change in the daily interactions of the classroom. Curriculum development traditionally refers to the changing, developing or reconstruction of an official (national or state) curriculum. Whilst Pinar et al. (1995) argue that the era of curriculum development has been replaced by an era of ‘curriculum understanding’, the so-called ‘curriculum developments’ of the 1980s and 1990s in New Zealand are discussed here. To effect a change in the students’ received curriculum requires a change in the teachers’ intended and taught curriculum, giving rise to the notion of curriculum development as teacher development (Bell, 1984b, 1986). This notion highlights that teacher development is usually required before any difference in the students’ received curriculum is noticed by students – this being the prime goal of curriculum developers.
Teacher development Teacher development may be viewed as the learning by teachers, in this case their learning to implement new curriculum policy or to use a new teaching or assessment strategy in the classroom. Changing the teachers’ intended and actual curriculum in the classroom, the assessed curriculum and the students’ constructed curriculum usually involves teacher learning, that is teacher development. The Learning in Science (Teacher Development) Project (Bell, 1993c) (see Chapter 1 for details of the research design and methodology) researched the learning of 34 primary and secondary teachers of science to use the research findings from the three previous Learning in Science Projects on teaching, learning and assessment of science and, in particular, teaching that takes into account students’ thinking (Bell, 1993b, d, 1994b). The often spoken phrase: ‘I know about LISP, but how do I put it into practice?’ was the centre of the research focus.
182
Science curriculum development and teacher development
The data generated by the teachers, engaging in teacher development in the research project, were analysed to give an overview of the adult learning process as it relates to the teachers’ learning or teacher development in the study. It was possible to describe three main types of development for the teachers involved in the research – social, personal and professional development (Bell and Gilbert 1994, 1996). Social development as part of teacher development involves the renegotiation and reconstruction of what it means to be a teacher (of science, for example). It also involves the development of ways of working with others that will enable the kinds of social interaction necessary for renegotiating and reconstructing what it means to be a teacher of science. For example, to use more open-ended investigations, a teacher might join a local group writing new units of work that include open-ended investigations. The group discussions on different teaching activities may be of help to a teacher concerned about the changed role of the teacher from one where the teacher is directing the practical work to one where the teacher is responding more to students’ initiatives. Personal development as part of teacher development involves each individual teacher constructing, evaluating and accepting or rejecting for herself or himself the new socially constructed knowledge about what it means to be a teacher (of science, for example), and managing the feelings associated with changing their activities and beliefs about science education, particularly when they go ‘against the grain’ (Cochran-Smith, 1991) of the current or proposed socially constructed and accepted knowledge. For example, a teacher using more open-ended investigations for the first time may have to manage feelings of agitation about the increased complexity and busyness of the classroom when the students are doing different activities rather than the same one, and their concerns about how the noisier classroom might be interpreted by other staff members. Professional development as a part of teacher development involves not only the use of different teaching activities but also the development of the beliefs and conceptions underlying the activities. It may also involve learning some science. For example, in addressing the use of more open-ended investigations, teachers may be required to rethink their ideas about the aims of science education, the nature of science, the role of the science teacher and classroom management. The model of teacher development A model was developed to describe and explain the teacher development that occurred as part of the research. A diagrammatic representation of the model is given in Figure 9.1. Initial personal development As part of this initial personal development, a teacher was aware, however inchoately, and accepting of a professional dissatisfaction or problem. Alternatively,
Science curriculum development and teacher development
Social development
Professional development
Social development 1: Seeing isolation as problematic Professional development 1: Trying out new activities Social development 2: Valuing collaborative ways of working and reconstructing what it means to be a teacher of science
Professional development 2: Development of ideas and classroom practice
Social development 3: Initiating collaborative ways of working
183
Personal development
Personal development 1: Accepting an aspect of my teaching as problematic
Personal development 2: Dealing with restraints
Personal development 3: Feeling empowered Professional development 3: Initiating other development activities
Figure 9.1 A model of teacher development
she or he may be wishing to be innovative and address what she or he perceives as a challenge rather than a problem. This development was usually private, having been self-initiated and sustained before the teacher engaged with the teacher development on offer. Prior to the teacher development programme, the teachers had decided that the broad type of activity might help overcome the dissatisfaction or problem or to take on a new challenge, and the risk of joining the group and the programme had been taken. Hence, the teachers saw the development activities as providing opportunities for their self-initiated growth, overcoming a professional problem, exploring an interesting avenue or helping implement new policy.
184
Science curriculum development and teacher development
A few of the teachers who joined the teacher development programme (as ‘volunteers’) had not undergone the personal development as described above. Therefore, whilst they attended the programme at the onset, they were not necessarily engaged in the learning, change and development initially. They may have been nominated for the programme by the school management, rather than the initiative coming from them; they may have gone along with peer pressure, for example when the rest of the science department volunteered and they didn’t want to be on the outside of the group. In these instances the individual personal development had not occurred before the programme commenced and their personal development needs had to be addressed within the programme, by both the facilitator and the other teachers. This usually involved helping the teachers to value their teaching competence overall, and to view only one aspect, that which was the focus of the programme, as problematic. No progress was made until this personal development was undertaken. Initial social development Before joining a teacher development programme, the teachers had also become aware that their isolation in the classroom was problematic. While being the only adult in a classroom can feel safe from negative criticisms and pressures to change, it does not always provide the new ideas, support and feedback necessary for teacher development (Hargreaves, 1992). In addition, while joining a teacher development programme involved taking some risks, the expected benefits of working with other teachers to improve teaching and learning were perceived as greater. The opportunities to discuss their teaching with other teachers and to renegotiate collectively what it means to be a teacher of science were seen positively and as helpful for change. Initial professional development The teachers appreciated clarifying the problematic aspect of their teaching. But while they appreciated clarifying an aspect that is problematic, they also needed to feel that their teaching overall was not problematic. The facilitator helped by communicating that they were perceived as competent teachers who were developing, rather than as teachers who were struggling. Valuing their ideas about teaching, by giving time in the sessions for them to talk about what they were doing in the classroom, was part of this communication. The teachers were encouraged to adopt the role of teacher-as-researcher. They valued finding out more information from their students and about their teaching; for example, what views of floating and sinking the students had and how many times in a lesson they asked the girls and the boys to answer their questions. This additional information helped them to clarify the problem they may have perceived in their teaching.
Science curriculum development and teacher development
185
The teachers were also asked to adopt the role of teacher-as-learner so that they viewed their professional development as learning rather than a remedial process. A supportive atmosphere helped to ease the uncomfortable feelings associated with learning – feeling incompetent or inadequate – and the uncomfortableness of getting in touch with feelings associated with prior experiences and beliefs. Part of the programme was given to enabling the teachers to learn about the change process. Gaining this metacognition was supportive in that it helped the teachers to understand what was happening for them. From the beginning of the programme, the teachers were given new teaching activities (Bell, 1993b) to use in the classroom, with the expectation that they would use the activities when they felt ready, and that they would have an opportunity to talk about their use of them in a sharing session. Using the new activities required prior planning; visualising what it might be like to use the activity; preparation of new resources; and being convinced that the new activity was needed and would work. The activities were small in scope – an activity not a unit of work or whole teaching sequence; were short in duration, for example, 10–30 minutes in length; could be done with a wide range of students, for example, across age and attainment groups; were not part of the official teaching and therefore could be done with a small group of students, rather than the whole class; and involved a change of teacher and student performance. Another important feature of the new teaching activities was that they were seen as likely to lead to better learning conditions, better classroom management, ‘feeling better about myself as a teacher’, and to better learning outcomes (Bell, 1993c: 154–214). The new activities could be understood, talked about with colleagues, and reflected on. However, the activities were also able to be used by the teachers as technicians and as novices rather than experts. Examples of the activities used were an interview-about-instances on burning and a survey based on a piece of research on students’ alternative conceptions of burning (Biddulph, 1991). The complete set of activities is given in Bell (1993b). These activities helped the teachers to find out more about their students’ thinking. In doing so, they were clarifying a problematic aspect of their teaching, adopting the role of teacher-as-researcher, and using a new teaching activity – one that can be used by a teacher who takes into account students’ thinking. The use of new teaching activities led to talking and thinking about the new teaching activities. The teachers wanted to talk about how the activities went for them. Initially, the discussions were about the concerns the teachers had with getting the activities to work with respect to classroom management and resource management. Later on, the discussions tended to be more about the educational issues involved in using the activities – for example the issue of assessment. The other teachers valued listening to whether the activities worked in the classroom, and discussing concerns and problems arising from their use. The talking went on in the sharing sessions through the telling of anecdotes (Bell, 1994c). The discussion arising helped the teachers to clarify their existing ideas
186
Science curriculum development and teacher development
on teaching and the role of the teacher, learning and learners. It also enabled new theoretical perspectives on science education to be introduced. For example, the findings of the interviews and surveys about burning (Biddulph, 1991) were shared with the rest of the group in the sharing sessions and the facilitator’s questions helped the teachers to clarify their ideas about ‘better’ learning (Bell and Pearson, 1992a). The notions of children’s science and constructivist views of learning were also introduced and reflected on over many sessions. In addition, the discussions raised the problem of what to do next; for example, how does the teacher interact with a student who views burning as a process in which things disappear? Problematic aspects of teaching were able to be clarified. A priority for the teachers was the development of a supportive atmosphere – one in which they felt encouraged to use the new activities; felt that their knowledge and expertise were valued and were seen as useful contributions; felt that their concerns about the possibility of judgements and put-downs were allayed; perceived that the feedback given was supportive and helpful; were able to share their problems and concerns publicly; felt supported; and found that their feelings associated with change were attended to in a non-threatening way. The second personal development As the teacher development continued, the teachers developed further in a personal way. Each individual teacher had to construct and evaluate for herself or himself an understanding of the socially reconstructed knowledge of what it means to be a teacher of science. This second phase of personal development also involved dealing with restraints; in particular, attending to the feelings and concerns of behaving differently in the classroom and changing their ideas about what it means to be a teacher of science. Their concerns included fear of losing control in the classroom; amount of teacher intervention; covering the curriculum; knowing the subject; meeting assessment requirements; relationships with students; and appraisal. These are discussed in greater detail, with supporting data, in Bell, 1993c; Bell and Gilbert, 1994, 1996. Addressing and resolving these concerns had both a cognitive and an affective strand. It appeared to the researchers that it was most crucial to address the affective dimensions if teacher development was to continue. Moreover, the development was both personal and social, in that the culture of what it means to be a teacher was being challenged and renegotiated by the group. Each individual teacher was having to position herself or himself with respect to the newly reconceptualised culture. The extent to which teachers were able to do this determined their level of engagement in the change process. In the programmes run as part of the research, these concerns or restraints were attended to in the sessions. First, the facilitator attempted to communicate that these restraints were concerns to be attended to and that teachers’ expressions of concern were not being viewed as giving excuses for not changing.
Science curriculum development and teacher development
187
Secondly, the teachers received suggestions from the facilitator and the other teachers on ways to get around the restraints. For example, one teacher shared how she got the students to compare their before and after concept maps with the learning objectives in the national curriculum for that topic. The students were able to give feedback to the teacher that they felt the curriculum content had been covered in their learning activities. At times, the concerns were addressed in a specific workshop activity. However, most of these concerns were addressed when they arose in the telling of anecdotes in the sharing sessions. The second social development As the programme continued, the teachers’ comments indicated that they were valuing collaborative ways of working. As the trust, support and credibility of the facilitator and other teachers became established, the teachers felt more able to contribute and more comfortable with contributing to the programme activities. They were more likely to share with the group anecdotes about what was happening in their classrooms; to give support and feedback to other teachers; to offer suggestions for new teaching activities; to suggest solutions to problems; and to voice their opinions and views. They were contributing to developing and sustaining collaborative relationships, which they valued. The collegial relationships were important as they provided opportunities for listening, contributing, discussing, supporting, giving feedback and reflecting on their teaching. In doing so, the teachers were renegotiating and reconstructing their shared knowledge about what it means to be a teacher of science. The second professional development The teachers continued to develop their ideas about science, science education and professional development and to develop their classroom activities. They were engaging in the development both of their thinking and of their classroom practice. With respect to their cognitive development they were: •
• • • • • •
clarifying their existing concepts and beliefs about science education – teaching, learning, the roles of the teacher, learners, the curriculum, the nature of science; obtaining an input of new information by listening and reading; constructing new understandings by linking the new information with existing ideas; considering, weighing up and evaluating the newly constructed understandings; accepting or rejecting the new constructions; using newly accepted understandings in a variety of contexts and with confidence; reconstructing what it means to be a teacher of science.
188
Science curriculum development and teacher development
With respect to the development of their classroom practice, they were: • • • • • •
obtaining new suggestions for teaching activities; considering them, visualising and planning for their use in the classroom; adapting and using the new activities; sharing their classroom experiences with others and obtaining feedback about the use of the activities; evaluating the new teaching activities; receiving support.
In particular, the teachers were more able to go beyond their classroom management of the new activities, and develop new ways of interacting with the students’ thinking. The development of the teachers’ classroom activities in this phase was usually to make use of the opportunities created by the use of the new activities to interact with the students’ thinking. The teachers were changing from being technicians using new activities to teachers who had a constructivist view of learning and who took into account students’ thinking. They were not only able to use the new activities from a classroom management point of view but were also able to respond to and interact with the students’ thinking. Being a teacher who has a constructivist view of learning was becoming a way of thinking and behaving for the teachers, rather than the implementation of some new teaching activities. They were trialling activities and thinking in the classroom to reconstruct what it means to be a teacher of science. The third personal development Towards the end of the programme, the teachers’ comments indicated that they were feeling more responsible for their own development. Developing a sense of trust is part of this personal development. The teachers appeared to trust the teaching activities to ‘work’, and to improve learning; with respect to the students they were more trusting that the students would have ideas on the topic being taught in science, that they would contribute to the discussions in the lesson, that they would learn the content of the curriculum, and that they would continue to take responsibility for their own learning. The teachers had to develop personally to be able to stand back and let go. There was personal development in attending to the feelings of apparent loss of control and not being centre-stage. Some of the teachers had to consider other kinds of feedback, for example indicators of learning, to maintain their sense of self-worth as a teacher. The third social development As the teachers developed more, they began actively to seek and initiate the activities and relationships with other teachers that they felt fostered their own
Science curriculum development and teacher development
189
development, for example asking a colleague for help; forming a network of colleagues; writing a unit of work for the school scheme or a teachers’ guide on one topic; and applying for contract jobs to facilitate regional teacher development programmes for the Ministry of Education. The third professional development Some of the teachers took initiatives to continue their development after the end of the programme run as part of the research project, by doing some form of curriculum development or facilitating teacher development programmes themselves. For example, some became involved in writing a unit of work for the school scheme or a teachers’ guide on one topic. Others introduced a colleague to the new teaching activities and facilitated the use of activities by the colleague in his or her classroom. Some applied for contract jobs to facilitate regional teacher development programmes for the Ministry of Education. They viewed writing and facilitating as additional ways to receive new theoretical and teaching ideas, support, feedback and further opportunities for talking with other teachers and for reflection. While they were giving, the teachers valued what they were receiving from the teachers with whom they were working. They were using the reconstructed knowledge about being a teacher of science. Evaluation of the teacher development The actual teacher development that occurred for the teachers involved in the Learning in Science (Teacher Development) Project was researched (Pearson and Bell, 1993). During the 1990–91 teacher development programmes, selfreport data was collected using surveys of teachers’ views of learning and learners, teaching and the role of the teacher, the nature of science and the role of science in the curriculum at the beginning and end of the teacher development programme in 1990, and at the end of 1991. The data indicated that the teachers had developed their views of teaching and learning, with positive changes in teacher beliefs and reported changes in classroom practice. In 1992, data was collected from another group of teachers undertaking the teacher development programme, using semi-structured pre- and post-interviews; pre- and post-surveys; and pre- and post-classroom observation of teachers, involving audiotape recordings and field notes. The self-report data from the surveys and interviews was similar to that obtained from the 1990–91 teachers: the teachers had developed their beliefs about teaching and learning. The preand post-observations indicated that all teachers had changed their teaching practice. The teachers were all using the activities suggested and modelled in the programme and which created the opportunities for the teachers to interact with the students’ thinking. In addition, seven of the ten teachers were observed using techniques involved in finding out students’ ideas and working with these ideas with the intent to promote conceptual development (Pearson and Bell, 1993).
190
Science curriculum development and teacher development
Summary The key points on teacher development from the Learning in Science Project (Teacher Development) are as follows: • • •
•
•
•
Teacher development can be viewed as teachers’ learning. The teacher development of the teachers in the research project can be described as social, personal and professional development. Teacher development can be seen as having two elements. One is the input of new theoretical ideas and new teaching suggestions. The second element is trying out, evaluating and practising these new theoretical and teaching ideas over an extended period, and in a collaborative situation where the teachers are able to receive support and feedback, critically reflect, and renegotiate and reconstruct what it means to be a teacher of science. In our experience, this second element tends to be underplayed in many in-service programmes and tends to occur through more informal modes such as telephone conversations, conversations in the staffroom, sharing anecdotes and visiting each other’s classrooms. Both elements are important if all three aspects of teacher development – social, personal, and professional – are to occur. The teachers’ development was occurring within the context of the effective components of a teacher development programme. These effective components were support, feedback and reflection (Barnett and Bell, 1997; Bell and Gilbert, 1996; Bell and Pearson, 1992b) and not an overall specified programme such as the particular in-service programme run in any one year of the research. The data reported here describe the learning process of the teachers, not a particular programme. Teacher development was helped when the teachers were able to talk with each other about what they were doing in the classroom, as an integral and key part of the programme. For example, the sharing sessions were structured around the use of anecdotes (Bell, 1994c). It was not something to be left to chance before or after any meetings. Taylor (1991) suggested that ‘radical pedagogical reform might require teachers to engage in the renegotiation of the culture of teaching, rather than going it alone’. The social development is seen as necessary for this renegotiation. Through talking with other teachers, the culture of teaching for the teachers in the study was being renegotiated (Tobin, 1993). The social communication and interaction among the teachers were important in the teacher development process. The evaluations indicated that the teachers changed their ideas about teaching and learning; used the activities in the programme to create the opportunities for teaching that takes into account students’ thinking; and some interacted and communicated with students in the teaching opportunities created to promote conceptual development.
Science curriculum development and teacher development
191
Curriculum development as teacher development In the previous section, teacher development was described as an aspect of curriculum development. The interdependence of curriculum development and teacher development has been considered in detail by Bell and Baker (1997), a summary version of which forms this section. In New Zealand in the 1990s, most of the teacher development activities were concerned with implementing the then new national science curriculum (Ministry of Education, 1993b). This teacher development focus was seen as one of compliance and involved teachers planning, teaching and assessing the new curriculum, and students experiencing and learning a new curriculum. In other words, after the rewriting of a national curriculum statement, each of the other multiple layers of the ‘curriculum’ (the planned, taught, learnt, assessed and hidden curricula) needed to be changed also. To help bring about these changes, many teachers were engaged in some form of teacher development (Gilmore, 1994) and had available teacher guides to implement the new curriculum (Ministry of Education, 1995a, 1995b). In other words, teacher development was a part of the implementation and compliance process and hence a part of curriculum development in the wider sense. But, as previously discussed, teacher development can be seen as a form of learning and human development involving professional, social and personal growth. Hence, teacher development is not just as an activity for teachers who need to ‘catch up’ with the latest developments but is an inherent part of being a professional who is developing with time and experience. Curriculum change relies on teachers and the factors which can be seen to promote teacher development for curriculum implementation are of interest. These are as follows. Curriculum implementation: professional development Professional development is required for curriculum implementation, and professional development is the main outcome sought by teachers, although social and personal development are crucial to achieving it. Some of the teacher development programmes contracted to the New Zealand Ministry of Education to implement the new science curriculum in the 1990s consisted of three one-day meetings, along with school visits by the facilitator between the meetings. From their experiences of the programme within this time frame, most teachers felt they understood the philosophy of the new curriculum but wanted more time and support to put the new ideas into practice (Gilmore, 1994). Curriculum implementation: social development The social development aspect presents a major challenge to teachers, teacher educators, school managers and policy makers as it requires the renegotiation and reconstruction of what it means to be a teacher of science as a key part of
192
Science curriculum development and teacher development
teacher development for curriculum implementation. There is also a challenge to empower teachers to be key contributors in the reconstruction of what it means to be a teacher of science. This includes being a contributor to the writing of national, school and classroom curricula; having a say in the nature and focus of teacher development programmes; and being in a position to change interrelated aspects of schooling, rather than just one aspect in isolation. Curriculum implementation: personal development Personal development is the third aspect of teacher development for curriculum implementation. Providing the support for the personal development aspect of teacher development is also a challenge for successful curriculum implementation. The stressful side of changing classroom activities while maintaining classroom control, maintaining learning outcomes, covering the curriculum and explaining the changes to students, parents and school management has been documented (Bell and Gilbert, 1996). The challenge is for provision in teacher development programmes (that is, more time and funding) for workshops on stress management, the change process and communication skills. The social–personal–professional view of teacher development is a challenge to the notion of curriculum development as a distinct process occurring prior to teacher development or implementation, which is the notion underlying the research, development, dissemination (RDD) model. In the social–personal–professional view, curriculum development is seen as a part of teacher development and teacher development is seen as a part of curriculum development – the two are concurrent and reciprocal. Changes to one aspect of classroom teaching by a teacher require changes to other aspects of teaching, schooling and education. For example, using a new teaching activity may require changing the assessment procedures in the classroom and school or the rewriting of the school curriculum (within the guidelines of a national curriculum). Whilst the literature acknowledges the role of teacher development in the implementation of new curricula, the role of curriculum development as a part of teacher development is not so well acknowledged. This view of teacher development then challenges the notion of curriculum development explicit in neo-conservative government policies, which separates out policy making from policy implementation, and which contracts out curriculum development tasks to be done over a short time frame. A reciprocal and interdependent view of curriculum development and teacher development has implications for the four levels at which curriculum and teacher development occurs. First, there is the national level, which involves curriculum and teacher development in the process of the development of a national curriculum (in science). The discussions to reach a negotiated document require the professional development of all stakeholders involved, including teachers, teacher educators, politicians, parents, scientists, employers and bureaucrats.
Science curriculum development and teacher development
193
Secondly, there is the school level of curriculum and teacher development. This may involve, for example, teachers undertaking teacher development for school change, such as developing the ability to write a school policy statement on equity or assessment. It may involve rewriting the school programme for science in the light of new national curricula. The teacher and curriculum development is intertwined. Thirdly, there is the level of change to a teacher’s curriculum in the classroom. Individual teachers may change their activities to implement a new school policy, new national curriculum policy or new innovations arising from research. This entails the teacher appraising their personal commitments to the school, the subject, their colleagues and students, to the proposed change, and a consideration of what the changes would mean within the (probably unchanged!) physical resources available. Lastly, there is the level of curriculum change as students experience it. Every student builds up a view of what the curriculum being experienced in a given subject is about. This is constructed from a personal overview of the whole school curriculum and the inter-relationship between its parts, from personal experience of the subject in previous years, from the opinions of peers, older siblings, parents or guardians, and from textbooks. No new teacher-intended curriculum should be introduced without an evaluation of the existing studentexperienced curriculum because curriculum development and teacher development are only effective if the students’ experienced curriculum changes in a way that improves learning. In a review covering the 1970s onward, Fullan and Hargreaves (1992) pointed out that, in all reported cases of successful curriculum development (curriculum innovation), teacher development was a contributing factor. However, they do point out the limitations to those (relatively few) sustained successes. They all involved sustained input from highly trained and experienced facilitators, who are in very short supply. Given the need to undertake multiple innovations in a school at the same time, Fullan and Hargreaves suggested that a comprehensive model of teacher development, the key to successful curriculum innovation, must take into account the following four elements: first, the teacher’s purpose, that is, what the teacher is trying to achieve; secondly, the teacher as a person; thirdly, the social contexts within which a teacher actually works, both in terms of the neighbourhood and within the school itself; and fourthly, the culture of teaching within the school. The model of teacher development as social, personal and professional development (Bell and Gilbert, 1996: 15) subsumes these elements within an articulated whole.
Teacher development as curriculum development The interdependence of curriculum development and teacher development has also been considered, as in the phrase ‘teacher development as curriculum development’ (Bell and Baker, 1997). In the 1980s, the aims and activities that
194
Science curriculum development and teacher development
dominated the teacher development programmes in science were those that could be categorised as ‘practice-to-policy’, or the development of innovation. Traditionally in New Zealand, teachers have been involved in the writing of a new national curriculum statement (in science). That is, teachers’ professional ideas, beliefs and practices have been incorporated into national curriculum policy. For example, in the 1980s, the Curriculum Review in Science (Bell, 1990) teachers and others involved in science education were members of the Forms 1–5 Science Revision Committee, who wrote a draft national science syllabus for students aged 10–15. Others were members of the eight development groups which developed and trialled new teaching activities and professional ideas in the areas of learning in science (two groups); the role of practical work; girls and science; technology and the science curriculum; developing a local science curriculum and network; Forms 1 and 2 science; and science education for Maori students. Their innovations were fed into the discussions and work of the Syllabus Revision Committee and recorded in a series of professional development guides (Ministry of Education, 1990b, 1990c, 1991a, 1991b, 1991c, 1991d). Teachers were also given the opportunity to respond to a number of discussion papers, for example (Department of Education, 1986a), newsletters (Department of Education, 1987–89) and a draft of the new Forms 1–5 Curriculum (Ministry of Education, 1990a). In the early 1990s national curriculum developments, teachers and others involved in science education were members of the writing team who wrote the actual official document; the reference groups who commented on the various drafts of the developing curriculum statement; and the ministerial Policy Advisory Committee who advised the Minister of Education on the national science curriculum policy (Bell, Jones and Carr, 1995; Haigh, 1995a). Teachers were also given the opportunity to respond to a draft of the new curriculum (Ministry of Education, 1992e). Hence, national curriculum development can be seen as involving ‘practiceto-policy’ activities, that is, the teachers’ ideas and practices were being used to develop and write new national curriculum policy. But in being involved in this curriculum development, the teachers are also learning new ideas and practices. The discussions, the trialling and testing out of innovative classroom practices, critique and reflection contributes to teachers’ learning, that is, their teacher development (professional, personal and social development). Practice-to-policy activities as a part of curriculum development are also teacher development activities. If curriculum development and teacher development are seen as interrelated, then viewing teacher development as social, professional and personal development has implications for the development of curriculum policy at the school and national levels. The factors which can be seen to promote the three components of teacher development in the ‘practice-to-policy’ form of curriculum development will now be discussed. These factors will be discussed under the headings of innovation, forums and a transparent curriculum development process.
Science curriculum development and teacher development
195
Innovations Teachers innovating in their classrooms can be seen as a key part in the development of new curriculum policy and when teachers are innovating in their classrooms, professional, social and personal development are involved. Teachers who participate in collaborative, action research and qualitative research also report on the teacher development that occurs as a result of their involvement in ongoing research (Bell and Cowie, 1997b; Kirkwood, 1988). In the 1990s, under the restructured education system in New Zealand, the funding for continued innovations in science education and education has become more insecure. In the past, innovations in classroom teaching have arisen from sustained research programmes – for example the Learning in Science Projects at the University of Waikato, and from ongoing curriculum development projects – for example the Curriculum Review in Science in the 1980s (Bell, 1990). The ways of changing official curriculum documents mean that opportunities for teachers to innovate, and develop professionally whilst doing so, were much reduced. The challenge to policy makers and funding agencies is to provide curriculum research and development funding for sustained, ongoing innovation which can contribute not only to the professional, social and personal development of the teachers innovating but can also inform future curriculum policy development. The Teaching and Learning Research Initiative in New Zealand (Ministry of Education, 2003c) is addressing this challenge in part. Taking risks is a key part of innovating in the classroom and personal development is involved. Teachers need a supportive climate for innovation and taking risks, especially the trialling of a new activity in the classroom. That climate does not exist at the current time when heavy demands are being made on teachers to implement new curricula, new prescriptions and new assessment strategies in the senior school and appraisal by performance criteria. Forums An implication for the social development aspect of teacher development for innovation and research is the provision of forums for debate for innovations and for future curriculum policy development, and for curriculum policy makers to be listening. There need to be appropriate forums for all the stakeholders in science education to negotiate and reconstruct aspects of science education during new policy development. Transparent processes An aspect of social, professional and personal development is teachers’ awareness of the processes of change and the processes of teacher development and curriculum development. Therefore, the process of national curriculum development needs to be accessible to teachers and educators, not inaccessible for those outside the policy organisations.
196
Science curriculum development and teacher development
National curriculum policy and development
Practice-to-policy
Research and development
Policy-to-practice Professional Personal Social
Teacher development
Professional Personal Social
Implementation
Curriculum -in-action
Figure 9.2 Teacher development and curriculum development
Summary The interdependence of curriculum development and teacher development can be summarised diagrammatically as in Figure 9.2, while keeping in mind that: • •
•
the curriculum is multi-layered and dynamic; curriculum development is not only the development of the official national curriculum but also the development of the school, planned, taught, learnt, assessed and hidden curricula; teacher development involves professional, personal and social development.
A number of points are highlighted by this diagrammatic summary. •
•
Teacher development is important in assisting teachers to contribute to the implementation aspect of curriculum development, for example in the implementation of new national curriculum policy. Teachers and teacher development are crucial if the curriculum, at all its levels, is to be effectively implemented. Teacher development is also important in assisting teachers to contribute to the curriculum development activities involving innovation and trialling
Science curriculum development and teacher development
•
•
•
197
leading up to the writing of a new national curriculum. Through their involvement in research, development and innovation, teachers can gain professional, personal and social development as well as contribute to future curriculum policy development. This teacher development is essential to ensuring that ongoing debates, innovation, curriculum research and development, teacher development and leadership are available for future curriculum policy development. Teachers are in an active position in the curriculum development processes, both implementation and research and development. These activities are legitimate professional activities for teachers who need to be involved. Curriculum development cannot be likened to a punctuated evolution, that is, significant change occurring with the release of a new national document, followed by a long period of little change. Rather, curriculum development is an ongoing evolution, with perhaps an increase in the rate when a new national document is developed. The ongoing development of teachers’ ideas and practices is important for any future development of a new curriculum statement. This teacher development will involve professional, personal and social development.
The shareholders in the New Zealand national science curriculum If the curriculum development process is seen as a political endeavour, then the people involved in negotiating a reconstructed curriculum are important to note. New Zealand has had a national science curriculum since the 1940s, with revisions or redevelopments occurring, on average, every 10–15 years (Bell and Baker, 1997). During each revision, the key groups who have an interest in science education in New Zealand schools debate and negotiate what knowledge and whose knowledge is valued as worthwhile to be in the national curriculum (Ministry of Education, 1993a) and the science national curriculum (Ministry of Education, 1993b). The key groups in New Zealand, who are shareholders in the science curriculum, include: early childhood, primary, secondary and tertiary teachers of science, students, science educators, Maori as Treaty partners, scientists including the Royal Society, parents and caregivers, employers, parliament politicians, and Ministry of Education officials. The contribution of four of these groups of shareholders, associated with the Waikato research, is now discussed: teachers, students, science educators and Maori. Teachers Since the 1970s, New Zealand primary and secondary science teachers have made an important contribution to the development of the national science curricula. Their views on what should be included in the curriculum have been considered,
198
Science curriculum development and teacher development
either by teachers on the national curriculum development team or through their feedback on a draft widely disseminated for comment before it became official (Bell, Jones and Carr, 1995). Their views have been considered alongside those of scientists and government officials (such as advisers and inspectors). Their views have added information to the debates about what science would be suitable for learning at different broad age bands; what science would be relevant, meaningful and engaging for the diverse groups of students; the ‘realities of the classroom’; the practicalities of what would ‘work’ in the context of a school laboratory, budgets, timetable and school management practices. Teacher views have also been indicative of how much change, at any one time, teachers of science could feasibly be expected to make to their teaching, i.e. the intended and actual curricula. These concerns have been reflected in the topics chosen by practising teachers for their masters and doctoral theses in science education at Waikato, for example meaningful contexts for girls’ learning of senior chemistry (Rodrigues, 1993b); meaningful contexts for Maori students of science (Gribble, 1993); meaningful contexts for learning senior physics (Porteous, 1997); ‘relevance’ in the curriculum development situation (Moli, 1993a, 1993b); critiques of new and existing curricula (Baimba, 1991; Ume, 1996); curricula to meet the needs of gifted students (Francis, 1998) and current teaching practices (Chang, 2000; Oxenham, 1995). Teacher/researcher contributions have also been made through teachers’ development of innovative teaching strategies as part of the action-research phases of the Learning in Science Projects for example (Kirkwood and Carr, 1988b; Osborne and Biddulph, 1985b; Osborne, Freyberg and Tasker, 1982). In these researches, curriculum development is viewed as action research (Bell, 1990). Teacher/researcher contributions have also been made through teachers being involved in collaborative research projects (Bell and Cowie, 1999; Bell and Gilbert, 1996). In addition, the teacher/researcher contributions have been published as teacher guide material produced by teachers as part of a research intervention, for example Cosgrove, Osborne and Forret, 1989; Hume, 1992; Kirkwood, 1989; Rodrigues, 1993a. In these research intervention and associated curriculum materials, curriculum development is viewed as implementing or using new learning and teaching approaches. In summary, teacher contributions to the development of new curricula include: • • •
their contributions to the writing of new curricula: the official and school curriculum; their implementation of new curricula: the teacher-intended and taught curricula; their development and use of innovative teaching and learning activities in response to eliciting the students’ learnt curriculum, for example during formative assessment.
Therefore, the teacher is a pivotal shareholder in any curriculum development.
Science curriculum development and teacher development
199
Students Few students have been involved in national curriculum debates and negotiations in New Zealand. Their voices, at times, have contributed to debates through others. For example, the research into students’ alternative conceptions has been considered, say in the case of ‘energy’ (Kirkwood and Carr, 1988a). Students’ views of classroom assessment have been researched (Cowie, 2000) but this work has had little impact to date in national science curriculum debates. The current Curriculum Stocktake publications (Carr et al., 2000; Hipkins et al., 2002; McGee et al., 2003; Ministry of Education, 2002) contain little of student voices. Science educators A shareholder group that has grown during the 1980s and 1990s is that of the science educators who are defined here as having a masters or doctoral qualification in science education. The group includes teachers, science teacher educators, for example in Colleges of Education and university Schools of Education; science education lecturers and researchers in university departments and research centres; advisers in School Support Services attached to teacher education institutions; Education Review Office officers and Ministry of Education officials. In 1979, when the first Learning in Science Project started, it was not possible to study masters papers in science education in New Zealand let alone obtain a masters degree in science education, as distinct from a masters in education or science. Doctorates in education, and in particular in science education, were few and far between. Indeed, the author was the first woman to get a doctoral degree in the field of education at the University of Waikato. This group of shareholders is notable as having been largely responsible for bringing the debates from the national and international science education research community into the national science curriculum development debates. Through this group, the research and development undertaken at Waikato was taken into account in subsequent curriculum developments, and in particular: • • • • • • •
the aims for science education of ‘science for all’ (Bell, 1988; Freyberg and Osborne, 1985); a constructivist view of learning, thinking and comprehension (Driver and Bell, 1986); teaching that takes into account students’ thinking (Barker, 1986; Biddulph, 1989; Cowie, 2000; Kirkwood, 1988); learning science in contexts meaningful for the students (Cosgrove, 1989; Cosgrove, Newman and Forret, 1987; Cosgrove and Meuggenberg, 1986); the inclusion of earth sciences (Happs, 1980b); Maori and science education (McKinley, 1998); the role of practical work (Haigh and Hubbard, 1997; Tasker and Freyberg, 1985).
200
Science curriculum development and teacher development
This is not a complete list as it contains only the Waikato contributions, given the topic of this book. Other New Zealand science educators have also contributed research and development findings to the curriculum debates, for example the many contributors in the book edited by Bell and Baker (1997). This group of shareholders, the science educators, has been involved in teacher development, including both pre- and in-service teacher education, for new and existing national science curricula, for example Fernandez (1991), Biddulph (1990a), Kirkwood (1988). It was within this group that the very visible public debate in New Zealand on constructivism arose (Matthews, 1993), accompanied by the publication of a book (Matthews, 1995) and its reviews (for example Bell, 1995b; Carr, 1995; Claxton, 1996; Gilbert, 1995; Jesson, 1995; McKinley, 1995a; Windross, 1995). The debates are an example of curriculum development as a political process, given that Matthews’ book was initiated and funded by the Business Roundtable. M3ori as Treaty of Waitangi partners As indicated in Chapters 7 and 8, Maori as partners to the Treaty of Waitangi are shareholders in the development of the New Zealand science national curriculum. The work of the Waikato group with respect to Maori and science curriculum development is documented in detail in Chapters 7 and 8.
Further developments The 1993 New Zealand science curriculum (Ministry of Education, 1993b) continues to be reviewed and revised. The development processes of the 1980s (Bell, 1990), of a syllabus revision group and development groups, and of the 1990s (Haigh, 1995a) of the writing group and reference groups, have been replaced by the processes of the New Zealand Curriculum Stocktake (Ministry of Education, 2002) and the New Zealand Curriculum Project (Ministry of Education, 2003b). However, these are best described as policy development rather than curriculum development, with the process being driven by policy analysts in the Ministry of Education. It is reminiscent of the top-down or centre–periphery modes of development, but within today’s managerial and economic discourses, with teachers’ work needing to be quantified, measured and controlled. However, for some, the term ‘curriculum development’ has been replaced by the term ‘understanding the curriculum’ (Pinar et al., 1995), where the curriculum is understood as various texts: understanding curriculum as historical text; political text; racial text; gender text; phenomenological text; poststructuralist, deconstructed, postmodern text; autobiographical/biographical text; aesthetic text; theological text, institutionalised text and as an international text. They argue that curriculum is ‘what the older generation chooses to tell the younger
Science curriculum development and teacher development
201
generation. . . . Curriculum becomes the site on which the generations struggle to define themselves and the world’ (Pinar et al., 1995: 847–8). The challenge they issue is for science education ‘to “take back” curriculum from the bureaucrats, to make the curriculum field itself a conversation, and in so doing to understand curriculum. We invite you to participate in that conversation’ (p. 848).
202
Glossary of M1ori words
Glossary of M1ori words
ako Aotearoa
pedagogy; teaching and learning the Maori name for New Zealand, literally ‘land of the long white cloud’ haka vigorous action dance hangi earthen oven, using heated stones and steam iwi tribe kai food kaimoana sea food kaitiaki guardian kaumatua male elder, old man or woman kaupapa collective wisdom, philosophy kaupapa Maori Maori philosophy kia ora a greeting of welcome kia piki ake i nga raruraru o te kainga socioeconomic mediation principle kiore a rat, resident in New Zealand before colonisation koha gift, donation kohanga reo Maori medium pre-school kuia respected woman elder, old woman, mother, grandmother kura kaupapa Maori Maori medium primary school mana status, prestige, dignity Maori the indigenous people of New Zealand marae meeting place maramataka fishing calendars matauranga Maori Maori knowledge nga mahi the work Pakeha the Maori name for New Zealanders of European descent Papatuanuku the earth
Glossary of M1ori words
Putaiao o Mataroa Ruamoko Taha Maori Tainui tangata whenua taonga taonga tuku iho tangi tapu te ao te ao Maori te ao marama te aho matua Te Arawa te kainga te kore te po te reo Maori te reo Maori me ona ahuatanga Te Tahuhu o te Matauranga Te Tauaki Marautanga Putaiao Te Taura Whiri i te Reo Maori Te Tiriti o Waitangi tikanga tikanga Maori tino Rangatiratanga tupuna Tuwharetoa waka waiata wairoa Waitangi whare wananga whakatauaki
203
science from the living world a Maori god of volcanoes and earthquakes literally, the Maori side; pertaining to Maori a group of iwi in the Waikato region indigenous people or literally ‘people of the land’ treasures cultural aspirations principle funeral sacred the world, daytime the Maori world after; the world of light, understanding and clarity parental vine, foundation document for kura kaupapa Maori a north island iwi the home before; nothingness during; night, place of departed spirits the Maori language the Maori language and its characteristics the (New Zealand) Ministry of Education the science curriculum the Maori language commission the Treaty of Waitangi customs Maori customs literally, chiefly control, Maori self-determination, relative autonomy ancestors a Waikato iwi canoe song waterway a district north of Auckland in the Bay of Islands place of learning proverbs
204
Glossary of M1ori words
whakapapa whanau whare kura Maori whare wananga Maori
genealogy extended family Maori medium secondary schools Maori tertiary institutions
References
205
References
Adey, P. (1997) It all depends on the context, doesn’t it? searching for general, educable dragons, Studies in Science Education, 29: 45–92. Aikenhead, G. (1996) Science education: border crossing into the subculture of science, Studies in Science Education, 27: 1–52. Aikenhead, G. (1997) Toward a first nation cross-cultural science and technology curriculum, Science Education, 81: 217–38. Apple, M. (1982) Education and Power. Boston: Routledge and Kegan Paul. Apple, M. (1996) Cultural Politics and Power. Buckingham: Open University Press. Appleton, K. (1984) Children’s Ideas about Hot and Cold. Working paper 127, Learning in Science Project (Primary). Hamilton: University of Waikato. Ash, D. and Levitt, K. (2003) Working within the zone of proximal development: formative assessment as professional development, Journal of Science Teacher Education, 14(1): 23. Assessment Reform Group (1999) Assessment for Learning: Beyond the Black Box. Cambridge: University of Cambridge. Atkin, J. M., Black, P., Coffey, J. and National Research Council (US) Committee on Classroom Assessment and the National Science Education Standards (2001) Classroom Assessment and the National Science Education Standards. Washington, DC: Center for Education National Research Council National Academy Press. Augoustinos, M. and Walker, I. (1995) Social Cognition: An Integrated Introduction. London: Sage Publications. Ausubel, D. (1968) Educational Psychology: A Cognitve View. New York, NY: Holt, Rinehart and Winston, Inc. Baimba, A. (1991) Innovations in science curriculum: a Sierra Leone case study. Unpublished PhD thesis, University of Waikato, Hamilton. Baird, J. (1984) Improving learning through enhanced metacognition. Unpublished PhD thesis, Monash University, Melbourne. Baird, J. and Mitchell, I. (eds) (1986) Improving the Quality of Teaching and Learning: An Australian Case Study – The PEEL Project. Melbourne: Monash University. Baird, J. and Northfield, J. (eds) (1992) Learning from the PEEL Experience. Melbourne: Monash University. Bakhtin, M. (1986) Speech genres and other late essays (trans. V. McGee) in C. Emerson and M. Holquist (eds) Austin, TX: University of Texas Press. Ball, S. (1990) Politics and Policy Making in Education. London: Routledge.
206
References
Barker, M. (1985a) Teaching and Learning about Photosynthesis. Working papers 220–9. Hamilton: University of Waikato. Barker, M. (1985b) Where Does Wood Come From? An Introduction to Photosynthesis for Third and Fourth Formers. A Teacher’s Guide. Hamilton: University of Waikato. Barker, M. (1986) The description and modification of children’s views of plant nutrition. Unpublished PhD thesis, University of Waikato, Hamilton. Barker, M. (1997) History in New Zealand science education: progress and prospects, in B. Bell and R. Barker (eds) Developing the Science Curriculum in Aotearoa New Zealand ( pp. 187–212). Auckland: Addison Wesley Longman New Zealand. Barker, M. (1999) The Maori language science curriculum in Aotearoa/New Zealand: a contribution to sustainable development, Waikato Journal of Education, 5: 51–60. Barker, M. (2001) How do people learn? Understanding the learning process, in C. McGee and D. Fraser (eds) The Professional Practice of Teaching (Second edition, pp. 35–66). Palmerston North: Dunmore Press. Barker, M. and Carr, M. (1989a) Photosynthesis – can our pupils see the wood for the trees?, Journal of Biological Education, 23(1): 41–4. Barker, M. and Carr, M. (1989b) Teaching and learning about photosynthesis. Part 1: An assessment in terms of students’ prior knowledge, International Journal of Science Education, 11(1): 49–56. Barker, M. and Carr, M. (1989c) Teaching and learning about photosynthesis. Part 2: A generative learning strategy, International Journal of Science Education, 11(2): 141–52. Barnett, J. and Bell, B. (1997) Supporting teacher development for curriculum implementation, in B. Bell and R. Baker (eds) Developing the Science Curriculum in Aotearoa New Zealand ( pp. 257–68). Auckland: Addison Wesley Longman Ltd. Barton, B. (1990) He Matauranga Tau Ahua Reorua: He Kitenga O Tetahi kaiako (Bilingual mathematics: a practitioner’s point of view). SAMEpapers 90: 159–77. Bell, B. (1981a) Towards Solutions: The Work of the Biology Action-research Group. Working Papers 29, 30, 31. Learning in Science Project. Hamilton: University of Waikato. Bell, B. (1981b) What is a plant? Some children’s ideas NZ Science Teacher, 31: 10–14. Bell, B. (1981c) When is an animal, not an animal? Journal of Biological Education, 15(3): 213–18. Bell, B. (1984a) The role of existing knowledge in reading comprehension and conceptual change in science education. Unpublished D.Phil thesis, University of Waikato. Bell, B. (1984b) The (UK) Secondary Science Curriculum Review, New Zealand Science Teacher, 42: 5–9. Bell, B. (1986) Form 1–5 Science Review – effecting change, New Zealand Science Teacher, 48: 6–9. Bell, B. (1987) Science curriculum development in New Zealand: an historical account, Research in Science Education, 17: 244–52. Bell, B. (1988) Girls and science, in S. Middleton (ed.) Women and Education in Aotearoa ( pp. 153–160). Wellington: Allen and Unwin, Port Nicolson Press. Bell, B. (1990) Curriculum development: a recent New Zealand example, SAMEpapers, 90: 1–31. Bell, B. (1991a) A constructivist view of learning and the draft F1–5 Science Syllabus, SAMEpapers, 91: 154–180. Bell, B. (1991b) Implications for curriculum, in J. Northfield and D. Symington (eds) Learning in Science Viewed as a Personal Construction: An Australian Perspective. Perth: SMEC Centre, Curtin University.
References
207
Bell, B. (1993a) Children’s Science, Constructivism and Learning in Science. Geelong, Australia: Deakin University Press. Bell, B. (1993b) Taking Into Account Students’ Thinking: A Teacher Development Handbook. Learning in Science Project (Teacher Development). Hamilton: University of Waikato. Bell, B. (1993c) (ed.) I Know about LISP, but How Do I Put It Into Practice? Final Report of the Learning in Science Project (Teacher Development). Hamilton: University of Waikato. Bell, B. (1993d) Teaching that takes into account students’ thinking. 1. Planning, New Zealand Science Teacher, 74: 4–7. Bell, B. (1994a) Guest editorial. Special issue on teacher development, International Journal of Science Education, 16(5): 493–5. Bell, B. (1994b) Teaching that takes into account students’ thinking. 2. Responding to the interacting with students’ thinking, New Zealand Science Teacher, 75: 21–5. Bell, B. (1994c) Using anecdotes in teacher development, International Journal of Science Education, Special Edition 1994, 16(5): 575–84. Bell, B. (1995a) Interviewing and assessment of science knowledge, in S. Glynn and R. Duits (eds) Learning Science in the Schools: Research Reforming Practice. Mahwah, NJ: Lawrence Erlbaum Press. Bell, B. (1995b) A response to M. Matthews (1995) Challenging New Zealand Science Education, in B. Bell (ed.) Responses to ‘Challenging NZ Science Education’ by Michael Matthews. Hamilton: University of Waikato. Bell, B. (1996) Teacher development in science education: policy-to-practice and practice-to-policy. Paper presented at the New Zealand Association for Research in Education Annual Conference, invited keynote address, 5–8 December, Nelson. Bell, B. (1998) Teacher development, in K. Tobin and B. Fraser (eds) International Handbook of Science Education (pp. 681–93). Dordrecht: Kluwer Academic Press. Bell, B. (2000) Formative assessment and science education: modelling and theorising, in R. Millar, J. Leach and J. Osborne (eds) Improving Science Education: The Contribution of Research. Buckingham: Open University Press. Bell, B. (2003) Theorizing pedagogy – Guest editorial, Waikato Journal of Education, 9: 3–9. Bell, B. (in press) Classroom assessment of science learning, in S. Abdell and N. G. Lederman (eds) The Handbook of Research on Science Education. Mahwah, NJ: Lawrence Erlbaum Associates. Bell, B. and Baker, R. (eds) (1997) Developing the Curriculum in Aotearoa New Zealand. Auckland: Addison Wesley Longman. Bell, B. and Barker, M. (1982) Towards a scientific concept of animal, Journal of Biological Education, 16(3): 197–200. Bell, B. and Brook, A. (1984) Aspects of Secondary Students’ Understanding of Plant Nutrition: Summary Report. Leeds: Children’s Learning in Science Project, University of Leeds. Bell, B. and Cowie, B. (1997a) Formative Assessment and Science Education. Summary Report of the Learning in Science Project (Assessment). Hamilton: University of Waikato. Bell, B. and Cowie, B. (1997b) Formative Assessment and Science Education. Research Report of the Learning in Science Project (Assessment). Hamilton: University of Waikato. Bell, B. and Cowie, B. (1999) Researching teachers doing formative assessment, in J. Loughran (ed.) Researching Teaching. London: Falmer Press. Bell, B. and Cowie, B. (2001a) The characteristics of formative assessment in science education, Science Education, 85(5): 536–53. Bell, B. and Cowie, B. (2001b) Formative Assessment and Science Education. Dordrecht; Boston: Kluwer Academic Publishers.
208
References
Bell, B. and Cowie, B. (2001c) Teacher development and formative assessment, Waikato Journal of Education, 7: 37–50. Bell, B. and Freyberg, P. (1985) Language in the science classroom, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implications of Children’s Science. Auckland: Heinemann. Bell, B. and Gilbert, J. (1994) Teacher development as professional, personal and social development, Teaching and Teacher Education, 10(5): 483–97. Bell, B. and Gilbert, J. (1996) Teacher Development: A Model from Science Education. London: Falmer Press. Bell, B., Jones, A. and Carr, M. (1995) The development of the recent national New Zealand Science Curriculum, Studies in Science Education, 26: 73–105. Bell, B. and Kirkwood, V. (1993) The teacher development programmes, in B. Bell (ed.) I Know about LISP but How Do I Put It Into Practice? Final report of the Learning in Science Project (Teacher Development). Hamilton: University of Waikato. Bell, B., Kirkwood, V. and Pearson, J. (1990a) Learning in Science Project ( Teacher Development). Framework. Working Paper 406. Hamilton: SMER Centre, University of Waikato. Bell, B., Kirkwood, V. and Pearson, J. (1990b) Learning in Science Project (Teacher Development). The Framework, Research in Science Education, 20: 31–40. Bell, B., Osborne, R. and Tasker, R. (1985) Finding out what children think, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implications of Children’s Science ( pp. 151–65). Auckland: Heinemann. Bell, B. and Pearson, J. (1991a) Changing. Working Paper 409, Learning in Science Project (Teacher Development). Hamilton: SMER Centre, University of Waikato. Bell, B. and Pearson, J. (1991b) Framework – 1991. Working Paper 410, Learning in Science Project (Teacher Development). Hamilton: SMER Centre, University of Waikato. Bell, B. and Pearson, J. (1991c) I Know about LISP but How Do I Put It Into Practice?, Research in Science Education, 21: 30–38. Bell, B. and Pearson, J. (1992a) ‘Better learning’, International Journal of Science Education, 14(3): 349–61. Bell, B. and Pearson, J. (1992b) Support in teacher development courses, SAMEpapers 92: 158–74. Bentley, D. and Watts, M. (1989) Learning and Teaching in School Science: Practical Alternatives. Milton Keynes: Open University Press. Benton, R. (1979) The MAori Language in the 1970s. Wellington: NZCER. Berger, P. and Luckmann, T. (1966) The Social Construction of Reality: A Treatise in the Sociology of Knowledge. New York, NY: Penguin Books. Berlak, H., Newmann, E., Adams, E., Archbald, D., Burgess, T., Raven, J. and Romberg, T. (eds) (1992) Toward a New Science of Educational Testing and Assessment. Albany, NY: State University of New York Press. Biddulph, F. (1983a) Primary School Children’s Ideas about Spiders. Working Paper 108, Learning in Science Project (Primary). Hamilton: University of Waikato. Biddulph, F. (1983b) Students’ Views of Floating and Sinking. Working Paper 116, Learning in Science Project (Primary). Hamilton: University of Waikato. Biddulph, F. (1984) Pupils’ Ideas about Flowering Plants. Working Paper 125, Learning in Science Project (Primary). Hamilton: University of Waikato. Biddulph, F. (1985) Using Children’s Questions in the Classroom: The Experience of Teachers and Pupils. Working Paper 133, Learning in Science Project (Primary). Hamilton: University of Waikato.
References
209
Biddulph, F. (1987a) Case Study in Teacher Change: An Intermediate School Syndicate Tries a Generative Approach in Science Teaching. Working Paper 402, Learning in Science Project (Teaching). Hamilton: University of Waikato. Biddulph, F. (1987b) Modelling Alternative Science Teaching Approaches for Pre-service Teachers. Working Paper 403, Learning in Science Project (Teaching). Hamilton: University of Waikato. Biddulph, F. (1989) Children’s questions; their place in primary science education. Unpublished PhD thesis, University of Waikato, Hamilton. Biddulph, F. (1990a) ‘I Still Have a Way To Go’: Case Study in Pre-service Change. Working Paper 407. Hamilton: SMER Centre, University of Waikato. Biddulph, F. (1990b) Pupil questioning as a teaching/learning strategy in primary science education, SAMEpapers 90: 60–73. Biddulph, F. (1991) Burning: A Science Unit for 10–13 Year Olds. University of Waikato, Hamilton. Biddulph, F. (1993) How do I know if I am a constructivist teacher? New roles for me as a teacher in the classroom: Appendix 1, in B. Bell (ed.) I Know about LISP, but How Do I Put It Into Practice? Final Report of the Learning in Science Project ( Teacher Development). Hamilton: University of Waikato. Biddulph, F. and Osborne, R. (1983) Children’s Ideas about Metals. Working Paper 112, Learning in Science Project (Primary). Hamilton: University of Waikato. Biddulph, F. and Osborne, R. (eds) (1984) Making Sense of Our World: An Interactive Teaching Approach. Hamilton: University of Waikato. Biddulph, F. and Osborne, R. (1985) Overviews: Selected Working Papers of the Learning in Science Project (Primary). Hamilton: University of Waikato. Biddulph, F., Osborne, R. and Freyberg, P. (1983a) The Framework Revisited. Working Paper 111, Learning in Science Project (Primary). Hamilton: University of Waikato. Biddulph, F., Osborne, R. and Freyberg, P. (1983b) Investigatory learning in science at the primary school level, Research in Science Education, 13: 223–32. Bishop, R. (2000) Nau te Rourou. Maori education: setting an agenda, Waikato Journal of Education, 6: 3–18. Bishop, R. and Glynn, T. (1999) Culture Counts: Changing Power Relationships in Education. Palmerston North: Dunmore Press. Bishop, R., Tiakiwai, S. and Richardson, E. (2003) Te kötahitanga: The Experiences of year 9 and 10 Maori Students Mainstream Classrooms. Report to the Ministry of Education. Hamilton: University of Waikato. Black, P. (1993) Formative and summative assessment by teachers, Studies in Science Education, 21: 49–97. Black, P. (1995) Assessment and feedback in science education, Studies in Educational Evaluation, 21(3): 257. Black, P. (2001) Dreams, strategies and systems: portraits of assessment past, present and furture, Assessment in Education, 8(1): 65–85. Black, P. and Harrison, C. (2001a) Feedback in questioning and marking: the science teacher’s role in formative assessment, School Science Review, 82(301): 55–61. Black, P. and Harrison, C. (2001b) Self- and peer-assessment and taking responsibility, School Science Review, 83(302). Black, P. and Wiliam, D. (1998a) Assessment and classroom learning, Assessment in Education, 5(1): 7–74. Black, P. and Wiliam, D. (1998b) Inside the black box – raising standards through classroom assessment, Phi Delta Kappan, 80(2) (Online).
210
References
Bol, L. and Strage, A. (1996) The contradictions between teachers’ instructional goals and their assessment practices in high school biology courses, Science Education, 80(2): 145–63. Brickhouse, N. (2001) Embodying science: a feminist perspective on learning, Journal of Research in Science Teaching, 38(3): 282–95. Broadfoot, P. (1996) Education, Assessment and Society. Buckingham: Open University Press. Brookhart, S. M. (2001) Successful students’ formative and summative uses of assessment information, Assessment in Education, 8(2): 153–69. Brown, G. and Desforges, C. (1977) Piagetian psychology and education: time for revision, British Journal of Educational Psychology, 47: 7–17. Bruner, J. (1977) The Process of Education. Cambridge, MA: Harvard University Press. Bruner, J. (1990) Acts of Meaning. Cambridge, MA: Harvard University Press. Bruner, J. and Haste, H. (eds) (1987) Making Sense: The Child’s Construction of the World. London: Routledge. Burden, P. (1990) Teacher development, in W. Houston (ed.) Handbook of Research on Teacher Education (pp. 311–28). New York, NY: Macmillan. Calabrese Barton, A. and Tobin, K. (2001) Urban science education, Journal of Research in Science Teaching, 38(8): 843–6. Carey, S. (1986) Cognitive science and science education, American Psychologist, 41(10): 1123–30. Carr, Malcolm (1991) Methods for studying personal construction, in J. Northfield and D. Symington (eds) Learning in Science Viewed as a Personal Construction. Perth: Curtin University of Technology. Carr, Malcolm (1995) Book review: Matthews, M. (1995) Challenging New Zealand Science Education. Palmerston North, Dunmore Press, Access, 13(2): 176–86. Carr, Malcolm., Barker, M., Bell, B., Biddulph, F., Jones, A., Kirkwood, V., Pearson, J. and Symington, D. (1994) The constructivist paradigm and some implications for science content and pedagogy, in P. Fensham and R. Gunstone and R. White (eds) The Content of Science: A Constructivist Approach to its Teaching and Learning (pp. 147–60). London: Falmer Press. Carr, Malcolm, and Kirkwood, V. (1988) Teaching and learning about energy in New Zealand secondary school classrooms, Physics Education, 23: 86–91. Carr, Malcolm, Kirkwood, V., Newman, B. and Birdwhistell, R. (1987a) Energy in three New Zealand secondary school junior classrooms, Research in Science Education, 17: 117– 28. Carr, Malcolm, Kirkwood, V., Newman, B. and Birdwhistell, R. (1987b) Energy in Three New Zealand Secondary School Junior Science Classrooms: An Overview. Working Paper 307, Learning in Science Project (Energy). Hamilton: University of Waikato. Carr, Malcolm, McGee, C., Jones, A., McKinley, E., Bell, B., Barr, H. and Simpson, T. (2000) Strategic Research: Initiative Literature Review: The Effects of Curricula and Assessment on Pedagogical Approaches and on Educational Outcomes. Wellington, New Zealand: Ministry of Education. Carr, Margaret (1997) Technological practice in early childhood as a dispositional milieu. Unpublished PhD thesis, University of Waikato, Hamilton. Carr, Margaret (2000) Technological affordance, social practice and learning narratives in an early childhood setting, International Journal of Technology and Design Education, 10(1): 81–101. Carr, Margaret (2001) Assessment in Early Childhood: Learning Stories in Learning Places. London: Paul Chapman.
References
211
Carr, Margaret, and Claxton, G. (2002) Tracking the development of learning dispositions, Assessment in Education, 9(9): 9–37. Chalmers, A. F. (1976) What Is This Thing Called Science? St Lucia, Queensland: University of Queensland Press. Chamberlain, M. (2001) The Development of Exemplars in New Zealand: Background and Rationale. Wellington: Ministry of Education, obtained from http://www.tki.org.nz/r/ assessment/two/research1_e.php, 30 October 2003. Chang, W. (2000) Improving teaching and learning of university physics in Taiwan. Unpublished PhD thesis, University of Waikato, Hamilton. Chang, W. and Bell, B. (2002) Making content easier or adding more challenge in Year One university physics, Research in Science Education, 32(1): 81–96. Chapman, M. (1986) The structure of exchange: Piaget’s sociological theory, Human Development, 29: 181–94. Charles, M. (1988) Teaching respiration to fourth formers. Unpublished MSc thesis, University of Waikato, Hamilton. Cheng, M. (1999) Teaching Resources Book – Heat. Hamilton: University of Waikato. Cheng, M. (2000) Redefining teacher education of primary teachers of science in Hong Kong. Unpublished PhD thesis, University of Waikato, Hamilton. Cheng, M. (2002a) A starting point of learning to teach science topics at the primary level – The Teacher Education Programme (original in Chinese), Asia-Pacific Forum on Science Learning and Teaching, http://www.ied.edu.hk/apfslt/, 2(2). Cheng, M. (2002b) Teaching Science with a Constructivist View of Learning – What Does it Mean in Practice? (original in Chinese). Hong Kong: Hong Kong Institute of Education. Cherryholmes, C. (1988) Construct validity and discourses of research, American Journal of Education, 96: 421–57. Christie, M. (1990) Aboriginal science for the ecologically sustainable future, Australian Science Teachers’ Journal, 37(1): 26–31. Chu, M. (1997) Asian-Chinese Students’ Learning in New Zealand secondary schools. Unpublished DPhil thesis, University of Waikato, Hamilton. Clarke, S. (1998) Targeting Assessment in the Primary Classroom: Strategies for Planning, Assessment, Pupil Feedback and Target Setting. London: Hodder and Stoughton. Clarke, S. (2001) Unlocking Formative Assessment; Practical Strategies for Enhancing Pupils’ Learning in the Primary Classroom. London: Hodder and Stoughton Educational. Clarke, S. (2003) Enriching Feedback in the Primary Classroom: Oral and Written Feedback from Teachers and Students. London: Hodder and Stoughton. Clarke, S., Timperley, H. and Hattie, J. (2003) Unlocking Formative Assessment: Practical Strategies for Enhancing Students’ Learning in the Primary and Intermediate Classroom. New Zealand Version. Auckland: Hodder Moa Beckett Publishers Ltd. Claxton, G. (1984) Teaching and acquiring scientific knowledge, in T. Keen and M. Pope (eds) Kelly in the Classroom: Educational Applications of Personal Construct Psychology. Montreal: Cybersystems. Claxton, G. (1991) Educating the Inquiring Mind: The Challenge for School Science. New York, NY: Harvester Wheatsheaf. Claxton, G. (1996) A SMER campaign: combating creeping constructivism, Studies in Science Education, 27: 157–65. Claxton, G. and Carr, M. (1987) Understanding Change: The Personal Perspective. Working Paper 401. Learning in Science Project (Teaching). Hamilton: University of Waikato.
212
References
Claxton, G. and Carr, M. (1991) Understanding change: the personal perspective, SAMEpapers 91: 1–14. Cobb, P. (1994) Where is the mind? Constructivist and sociocultural perspectives on mathematical development, Educational Researcher, 20(7): 13–20. Cochran-Smith, M. (1991) Learning to teach against the grain, Harvard Educational Review, 61(3): 279–310. Cognition and Technology Group at Vanderbilt (1990) Anchored instruction and its relationship to situated cognition, Educational Researcher, 19(6): 2–10. Collier, G., Minton, H. and Reynolds, G. (eds) (1991) Currents of Thought in American Social Psychology. New York: Oxford University Press. Cosgrove, M. (1982a) Mixtures. Working Paper 56. Learning in Science Project. Hamilton: University of Waikato. Cosgrove, M. (1982b) Mixtures: An Introduction to Chemistry. Hamilton: Hamilton Teachers College. Cosgrove, M. (1989) Learning science from technology. Unpublished DPhil thesis, University of Waikato, Hamilton. Cosgrove, M. and Meuggenberg, G. (1986) Learning Science from Technology. Refrigeration: A Teaching Unit for Form Six Physics. Hamilton: University of Waikato. Cosgrove, M., Newman, B. and Forret, M. (1987) Teaching technology – refrigeration, an example from New Zealand’s economic history, in K. Riquarts (ed.) Science and Technology and the Quality of Life: Proceedings of the 4th International Symposium on World Trends in Science and Technology Education, Vol. 2, pp. 467–73. Kiel: Kiel Institute for Science Education. Cosgrove, M. and Osborne, R. (1980) Physical Change. Working Paper 26, Learning in Science Project. Hamilton: University of Waikato. Cosgrove, M. and Osborne, R. (1985a) Lesson frameworks for changing children’s ideas, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implications of Children’s Science. Auckland: Heinemann. Cosgrove, M. and Osborne, R. (1985b) A teaching sequence on electric current, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implication of Children’s Science. Auckland: Heinemann. Cosgrove, M., Osborne, R. and Forret, M. (1989) Electric Current: Developing Learners’ Views: Teachers’ Guide, third edition. Hamilton: Waikato Education Centre. Cowie, B. (1997) Formative assessment and science classrooms, in B. Bell and R. Baker (eds) Developing the Science Curriculum in Aotearoa New Zealand. Auckland: Longman Addison Wesley. Cowie, B. (1998) The role of questions in classroom assessment, SAMEpapers, 98: 41–53. Cowie, B. (2000) Formative assessment in science classrooms. Unpublished PhD thesis, University of Waikato, Hamilton. Cowie, B. and Bell, B. (1995) Research Report 1: Views of Assessment. Learning in Science Project (Assessment). Hamilton: University of Waikato. Cowie, B. and Bell, B. (1996a) Towards a Model of Formative Assessment in the Classroom. Interim Research Report 2. Hamilton: University of Waikato. Cowie, B. and Bell, B. (1996b) Validity and formative assessment in the classroom. Paper presented at the invited keynote paper to the International Symposium on Validity in Educational Assessment, University of Otago, Dunedin, NZ, 28–30 June. Cowie, B. and Bell, B. (1999a) A methodology for investigating formative assessment in the science classroom. SAMEpapers, 99: 36–62.
References
213
Cowie, B. and Bell, B. (1999b) A model of formative assessment in science education, Assessment in Education, 6(1): 102–16. Cowie, B., Boulter, C. and Bell, B. (1996) Developing a Framework for Assessment of Science in Classrooms. A working paper of the Learning in Science Project (Assessment). Hamilton: Centre for Science, Mathematics and Technology Education Research, University of Waikato. Crooks, T. (1988) The impact of classroom evaluation practices on students, Review of Educational Research, 58(4): 438–81. Crooks, T. (2002a) Assessment, accountability and achievement – principles, possibilities and pitfalls. Paper presented at the keynote address to the annual conference of the New Zealand Association for Research in Education, Palmerston North, NZ, 5–8 December. Crooks, T. (2002b) Educational assessment in New Zealand schools, Assessment in Education, 9(2): 237, 217. Crooks, T. and Flockton, L. (1996) National Education Monitoring Report 1: Science Assessment Results 1995. Dunedin: Educational Assessment Research Unit, University of Otago. Crooks, T., Kane, M. and Cohen, A. (1996) Threats to the valid use of assessments, Assessment in Education, 3(3): 265–85. Cumming, J. and Maxwell, G. (1999) Contextualising authentic assessment, Assessment in Education, 6(2): 177–94. Darling-Hammond, L. (1994) Performance-based assessment and educational equity, Harvard Educational Review, 64(4). Davies, B. (1993) Shards of Glass. Sydney: Allen and Unwin. Delpit, L. D. (1993) The silenced dialogue: power and pedagogy in educating other people’s children, in L. Weis and M. Fine (eds) Beyond Silenced Voices: Class, Race and Gender in United States Schools (pp. 119–39). Albany, NY: Suny Press. Department of Education (1978) Science: Forms One to Four Draft Syllabus and Guide. Wellington: Department of Education. Department of Education (1979) Science Syllabus and Guide. Primary: To Standard Four. Wellington: Department of Education. Department of Education (1984) Taha Maori; Suggestions for Getting Started. Wellington: Department of Education. Department of Education (1986a) Curriculum Review in Science. A Discussion Paper 8: Multicultural Science. Wellington: Curriculum Development Division, Department of Education. Department of Education (1986b) Learning and Achieving: Second Report of the Committee of Inquiry into Curriculum, Assessment, and Qualifications in Forms 5 to 7. Wellington: Department of Education. Department of Education (1987–89) Curriculum Review in Science Newsletters. Wellington: Curriculum Development Division. Department of Education (1988a) Interactive Teaching in Science: Floating and Sinking. A Video. Auckland: Auckland College of Education Video Unit. Department of Education (1988b) Te Ao Turoa Reo Rua: A Bilingual Science Resource. Wellington: Curriculum Development Division. Ditchfield, C. (1987) Better Science: Working for a Multicultural Society. Curriculum Guide 7. Heinemann Educational Books/Association for Science Education for the School Curriculum Development Committee. Doll, R. (1989) Curriculum Improvement: Decision Making and Process. Boston: Allyn and Bacon. Donaldson, M. (1978) Children’s Minds. Glasgow: Fontana/Collins.
214
References
Doran, R. L., Lawrenz, F. and Helgeson, S. (1993) Research on assessment in science, in D. Gabel (ed.) Handbook of Research in Science Teaching and Learning (pp. 388–442). New York, NY: Macmillan. Driver, R. (1978) ‘When is a stage not a stage?’ A critique of Piaget’s theory of cognitive development and its implication to science education, Educational Researcher, 21(1): 54– 61. Driver, R. (1981) Pupils’ alternative frameworks in science, European Journal of Science Education, 3(1): 93–101. Driver, R. (1982) Children’s learning in science, Educational Analysis, 4(2): 69–79. Driver, R. (1983) The Pupil as Scientist? Milton Keynes: Open University Press. Driver, R. (1989) Children’s conceptions and the learning in science, International Journal of Science Education, 11(5): 481–90. Driver, R., Asoko, H., Leach, J., Mortimer, E. and Scott, P. (1994) Constructing scientific knowledge in the classroom, Educational Researcher, 23(7): 5–12. Driver, R. and Bell, B. (1986) Students’ thinking and the learning of science, School Science Review, 67(240): 443–56. Driver, R. and Easley, J. (1978) Pupils and paradigms: a review of literature related to concept development in adolescent science students, Studies in Science Education, 5: 61– 84. Driver, R. and Erickson, G. (1983) Theories-in-action; some theoretical and empirical issues in the study of students’ conceptual frameworks in science, Studies in Science Education, 10: 37–60. Driver, R., Guesné, E. and Tiberghien, A. E. (1985) Children’s Ideas in Science. Milton Keynes: Open University Press. Driver, R. and Oldham, V. (1986) A constructivist approach to curriculum development in science, Studies in Science Education, 13: 105–22. Driver, R., Squires, A., Rushworth, P. and Wood-Robinson, V. (1994) Making Sense of Secondary Science: Research into Children’s Ideas. London: Routledge. Duckworth, E. (1991) Twenty-four, forty-two, and I love you: keeping it complex, Harvard Educational Review, 61(1): 1–24. Duckworth, E., Easley, J., Hawkins, D. and Henriques, A. (1990) Science Education: A Minds-on Approach for the Elementary Years. London: Lawrence Erlbaum. Duit, R. (1994) Research on students’ conceptions – developments and trends, in H. Pfundt and R. Duit (eds) Bibliography: Students’ Alternative Frameworks and Science Education. Kiel: IPN. Dunkin, M. and Biddle, B. (1974) The Study of Teaching. New York, NY: Holt, Rinehart and Winston, Inc. Durie, M. (2001) Educational goals for Maori education. Paper presented at the Hui Taumata Matauranga, Taupo, February. Duschl, R. and Gitomer, D. H. (1997) Strategies and challenges to changing the focus of assessment and instruction in science classrooms, Educational Assessment, 4: 37–73. Eames, C. (2000) Learning in the workplace through co-operative education: beginning a longitudinal qualitative study, Journal of Cooperative Education, 35(2–3): 76–83. Eames, C. (2003a) Learning through co-operative education work placements in science and technology. Unpublished PhD thesis, University of Waikato, Hamilton. Eames, C. (2003b) Researching in cooperative education: how a practitioner met the challenge!, in P. Linn, A. Howard and E. Miller (eds) Handbook for Research in Cooperative Education and Internships. Mahwah, NJ: Lawrence Erlbaum Associates.
References
215
Edwards, D. (1991) Categories are for talking: on the cognitive and discursive bases of categorisation, Theory and Psychology, 1: 515–42. English, C. and Wood, R. (1997) The doing of science – a redefinition of procedural knowledge, in B. Bell and R. Baker (eds) Developing the Science Curriculum in Aotearoa New Zealand (pp. 67–80). Auckland: Addison Wesley Longman New Zealand. Evening, L. (1998) Exploration into the under-representation of girls in physics in Fiji secondary schools. Unpublished MPhil thesis, University of Waikato, Hamilton. Faire, J. and Cosgrove, M. (1988) Teaching Primary Science. Hamilton: Waikato Education Centre. Fensham, P. (1985) Science for all: a reflective essay, Journal of Curriculum Studies, 17: 415– 35. Fensham, P., Gunstone, R. and White, R. (1994) The Content of Science: A Constructivist Approach to its Teaching and Learning. London: Falmer Press. Fernandez, T. (1991) Implementing an alternative primary school science pedagogy: experiences of beginning teachers. Unpublished MEd thesis, University of Waikato, Hamilton. Fernandez, T. and Ritchie, G. (1992) Reconstructing the interactive science pedagogy: experiences of beginning teachers implementing the interactive science pedagogy, Research in Science Education, 22: 123–31. Filer, A. (ed.) (2000) Assessment: Social Practice and Social Product. London: RoutledgeFalmer. Filer, A. and Pollard, A. (2000) The Social World of Pupil Assessment: Processes and Contexts of Primary Schooling. London: Continuum. Flockton, L. (1999) School-wide Assessment: National Education Monitoring Project. Wellington: New Zealand Council for Educational Research. Flockton, L. and Crooks, T. (2000) Assessment Results for Maori Students, 1999: Science, Art, Graphs, Tables and Maps. Dunedin, NZ: Educational Assessment Unit, University of Otago. Fox, R. (2001) Constructivism examined, Oxford Review of Education, 27(1): 23–35. Francis, I. (1998) Towards better serving the needs of year 11 to 13 students considered gifted in physics in the New Zealand educational setting. Unpublished MEd, University of Waikato, Hamilton. Freyberg, P. and Osborne, R. (1981) Who structures the curriculum: teacher or learner?, SET, 2(Item 6). Freyberg, P. and Osborne, R. (1982) Final Report: Learning in Science Project (Form 1–4). Hamilton: University of Waikato. Freyberg, P. and Osborne, R. (1985) Assumptions about teaching and learning, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implications of Children’s Science. Auckland: Heinemann. Freyberg, P. and Osborne, R. (1985) Constructing a survey of ‘alternative’ views, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implications of Children’s Science (pp. 166–67). Auckland: Heinemann. Freyberg, P., Osborne, R. and Tasker, R. (1979) Problems and Difficulties: The Working Papers of the Exploratory Phase. Learning in Science Project (Forms 1–4). Hamilton: University of Waikato. Fullan, M. and Hargreaves, A. (eds) (1992) Understanding Teacher Development. New York, NY: Teachers’ College Press. Fusco, D. and Barton, A. C. (2001) Representing student achievements in science, Journal of Research in Science Teaching, 38(3): 337–54.
216
References
Fuss, D. (1989) Essentially Speaking: Feminism, Nature and Difference. New York, NY: Routledge. Gagné, R. (1970) The Conditions of Learning. New York, NY: Holt, Rinehart and Winston. Gauld, C. (1980) Subject orientated test construction, Research in Science Education, 10: 77–82. Gauld, C. (1986) Models, meters and memory, Research in Science Education, 16: 49–54. Gilbert, Jane (1993) Teacher development: a literature review, in B. Bell (ed.) I know about LISP, but How Do I Put It Into Practice? Final Report of the Learning in Science Project (Teacher Development). Hamilton: University of Waikato. Gilbert, Jane (1994) The construction and reconstruction of the concept of the reflective practitioner in the discourses of teacher professional development, International Journal of Science Education, 16(5): 511–22. Gilbert, Jane (1995) Feminism and science education: a response to M. Matthews’ ‘Challenging NZ Science Cirriculum’, in B. Bell (ed.) Responses to ‘Challenging NZ Science Education’. Hamilton: University of Waikato. Gilbert, Jane (1997) Thinking ‘other-wise’: re-thinking the problem of girls and science education in the post-modern. Unpublished DPhil thesis, University of Waikato, Hamilton. Gilbert, Jane (2001) Science and its ‘Other’: Looking underneath ‘woman’ and ‘science’ for new directions in research on gender and science education, Gender and Education, 13(3): 291–305. Gilbert, John, Osborne, R. and Fensham, P. (1982) Children’s science and its consequences for teaching, Science Education, 66(4): 623–33. Gilbert, John and Pope, M. (1986) Small group discussions about conceptions in science; a case study, Research in Science and Technology Education, 4(1): 61–76. Gilbert, John and Watts, M. (1983) Concepts, misconceptions and alternative conceptions: changing perspectives in science education, Studies in Science Education, 10: 61– 98. Gilbert, John, Watts, M. and Osborne, R. (1982) Students’ conceptions of ideas in mechanics, Physics Education, 17: 62–6. Gill, D. and Levidow, L. (1987) (ed.) Anti-racist Science Teaching. London: Free Association Books. Gilmore, A. (1994) Evaluation of the Teacher Development Programmes: Mathematics and Science in the National Curriculum. Christchurch: University of Canterbury. Gilmore, A. and Hattie, J. (2001) Understanding usage of an Internet-based information resource for teachers: the assessment resource banks, New Zealand Journal of Educational Studies, 36(2): 237–57. Gil-Pérez, D., Guisasola, J., Moreno, A., Cachapuz, A., Pessoa De Carvalho, A. M., Martínez Torregrosa, J. et al. (2002) Defending constructivism in science education, Science and Education, 11: 557–71. Gipps, C. (1994a) Beyond Testing: Towards a Theory of Educational Assessment. London: Falmer Press. Gipps, C. (1994b) Developments in educational assessment or what makes a good test?, Assessment in Education, 1(3). Gipps, C. (1998) Equity in education and assessment. Paper presented at the Keynote address to the annual conference of the New Zealand Association for Research in Education, Dunedin, December. Gipps, C. (1999) Socio-cultural aspects of assessment, Review of Research in Education, 24: 355–92.
References
217
Gipps, C. and James, M. (1998) Broadening the basis of assessment to prevent the narrowing of learning, The Curriculum Journal, 9(3): 285–97. Gipps, C. and Murphy, P. (1994) A Fair Test? Buckingham: Open University Press. Gipps, C. and Tunstall, P. (1996a) ‘How does your teacher help you to make your work better?’ Children’s understanding of formative assessment, Curriculum Journal, 7(2): 185– 203. Gipps, C. and Tunstall, P. (1996b) Teacher feedback to young children in formative assessment: a typology, British Educational Research Journal, 22(4): 389–404. Gipps, C. V. (1994) Beyond Testing: Towards a Theory of Educational Assessment. London: Falmer Press. Gitomer, D. H. and Duschl, R. (1998) Emerging issues and practices in science assessment, in B. J. Fraser and K. Tobin (eds) International Handbook of Science Education. Dordrecht: Kluwer Academic Publishers. Glynn, S. and Duit, R. (1995) Learning Science in the Schools: Research Reforming Practice. Mahwah, NJ: Lawrence Erlbaum Associates. Glynn, T. (1998) Bicultural challenges for educational professionals in Aotearoa, Waikato Journal of Education, 4: 3–16. Gore, J. (1993) The Struggle for Pedagogies: Critical and Feminist Discourses as Regimes of Truth. London: Routledge. Gore, J. (1998) Disciplining bodies: on the continuity of power relations in pedagogy, in T. Popkewitz and M. Brennan (eds) Foucault’s Challenge: Discourse, Knowledge and Power in Education ( pp. 231–51). New York, NY: Teachers’ College Press. Grant, P., Johnson, L. and Sanders, Y. (1990) Better Links: Teaching Strategies in the Science Classroom. STAV Publishing, Parkville, Victoria. Gribble, W. (1993) The use of Maori myths and legends as a teaching aid in earth sciences. Unpublished MSc, University of Waikato, Hamilton. Guba, E. G. and Lincoln, Y. S. (1989) Fourth Generation Evaluation. Newbury Park, CA: Sage Publications. Gunstone, R. (1988) Learners in science education, in P. Fensham (ed.) Development and Dilemmas in Science Education. London: Falmer Press. Gunstone, R. (1990) Children’s science: a decade of developments in constructivist views of science teaching and learning, Australian Science Teachers’ Journal, 36(4): 9–19. Haami, B. (1993) Cultural knowledge and traditions relating to the kiore rat in Aotearoa. Part 1: A Maori perspective, SAMEpapers, 93: 5–22. Haigh, M. (1993) ‘Hands on – minds-on’: introducing openness into senior biology practical work, Research in Science Education, 23, 110–17. Haigh, M. (1995a) The achievement initiative in science: Developing ‘Science in the New Zealand Curriculum’, SAMEpapers, 95: 45–74. Haigh, M. (1995b) Open Investigative Work in Level 7 Biology. Auckland: Auckland College of Education. Haigh, M. (1998) Investigative practical work in Year 12 biology programmes. Unpublished PhD thesis, University of Waikato, Hamilton. Haigh, M. (2000) The many faces of case study research as illustrated through case studies of investigative practical work in Year 12 biology programmes, SAMEpapers, 00: 69–93. Haigh, M. (2001) Supporting student investigators, New Zealand Science Teacher, 96: 39–41. Haigh, M. and Hubbard, D. (1997) ‘I really know I have learnt something’: investigative work in science education, in B. Bell and B. Baker (eds) Developing the Science Curriculum in Aotearoa New Zealand (pp. 53– 66). Auckland: Addison Wesley Longman New Zealand Ltd.
218
References
Hameed, H. (1997) Critical thinking dispositions: assessment and enhancement by infusion in physics. Unpublished PhD thesis, University of Waikato, Hamilton. Happs, J. (1980a) Particles. Working Paper 18, Learning in Science Project. Hamilton: University of Waikato. Happs, J. (1980b) Towards an earth science curriculum at the Forms 6 and 7 levels in New Zealand. Unpublished MSc thesis, University of Waikato, Hamilton. Happs, J. (1983) Using socio-conflict to establish an understanding of the scientific meaning of rock, Research in Science Education, 13: 61–71. Happs, J. (1984) The utility of alternative knowledge frameworks in effecting conceptual change; some examples from the Earth Science. Unpublished DPhil thesis, University of Waikato, Hamilton. Harding, S. (1991) Whose Science? Whose Knowledge? Thinking from Women’s Lives. Milton Keynes: Open University Press. Harding, S. (1993) The Racial Economy of Science: Towards a Democratic Future. Bloomington, IN: Indiana University Press. Hargreaves, A. (1992) Cultures of teaching; a focus for change, in A. Hargreaves and M. Fullan (eds) Understanding Teacher Development (pp. 216– 40). New York, NY: Teachers’ College Press. Hargreaves, A. and Fullan, M. (eds) (1992) Understanding Teacher Development. New York, NY: Teachers’ College Press. Harlen, W. (1995) To the rescue of formative assessment, Primary Science Review, 37: 14 – 16. Harlen, W. (1998) Classroom assessment: a dimension of purposes and procedures. Paper presented at the keynote address given to annual conference of the New Zealand Association of Educational Research, Dunedin, December. Harlen, W. (1999) Purposes and procedures for assessing science process skills, Assessment in Education, 6(1): 129. Harlen, W. and James, M. (1996) Creating a positive impact of assessment on learning. Paper presented at the AERA Conference, New York. Harlen, W. and James, M. (1997) Assessment and learning, Assessment in Education, 4(3): 365. Harlen, W. and Osborne, R. (1983) Toward a Teaching Model for Primary Science. Working Paper 114, Learning in Science Project (Primary). Hamilton: University of Waikato. Harré, R. (1986) Varieties of Realism. Oxford: Blackwell. Harré, R. and Gillett, G. (1994) The Discursive Mind. London: Sage Publications. Hattie, J. (1999) Influences on student learning. Paper presented at the inaugural professorial lecture, University of Auckland. http://www.arts.auckland.ac.nz/edu/staff/ jhattie/Inaugural.html. Hattie, J. (2003) asTTle: Assessment Tools for Teaching and Learning. Ministry of Education. Retrieved from the World Wide Web: http://www.tki.org.nz/r/asttle/ tech_model_3_e.php Hattie, J., Biggs, J. and Purdie, N. (1996) Effects of learning skills interventions on student learning: a meta-analysis, Review of Educational Research, 66: 99–136. Hattie, J. and Jaeger, R. (1998) Assessment and classroom learning: a deductive approach, Assessment in Education, 5(5): 111. Hawe, E. (1983) Pupils’ Views about Spiders. Working Paper 123, Learning in Science Project (Primary). Hamilton: University of Waikato.
References
219
Hemara, W. (2000) MAori Pedagogies: A View from the Literature. Wellington: New Zealand Council for Educational Research. Hennessy, S. (1993) Situated cognition and cognitive apprenticeship; implications for classroom learning, Studies in Science Education, 22: 1– 41. Hesse, M. B. (1966) Models and Analogies in Science. Notre Dame, IN: University of Notre Dame Press. Hewson, P. (1981) The conceptual change approach to learning science, European Journal of Science Education, 3(4): 383–96. Hewson, P. and Hewson, M. (1988) An appropriate conception of teaching science: a view from the studies of science learning, Science Education, 72(5): 597–614. Hewson, P. and Thorley, N. (1989) The conditions of conceptual change in the classroom, International Journal of Science Education, 11(special issue): 541–53. Hickey, D. T. and Zuiker, S. J. (2003) A new perspective for evaluating innovative science programs, Science Education, 87(4): 539–63. Hill, M. (1999) Assessment in self-managing schools: primary teachers balancing learning and accountability demands in the 1990s, New Zealand Journal of Educational Studies, 34(1): 176–85. Hill, M. (2000) Remapping the assessment landscape: primary teachers reconstructing assessment in self-managing schools. Unpublished PhD thesis, University of Waikato, Hamilton. Hills, G. (1989) Students’ ‘untutored beliefs’ about natural phenomena: primitive science or commonsense?, Science Education, 73(2): 155–86. Hipkins, R. and Arcus, C. (1997) Teaching science in context: challenges and choices, in B. B. F. and R. Baker (eds) Developing the Science Curriculum in Aotearoa New Zealand (pp. 101–14). Auckland: Addison Wesley Longman New Zealand. Hipkins, R. and Barker, M. (2002) Science in the New Zealand Curriculum: present, potential and future possibilities, New Zealand Science Teacher, 100: 10–17. Hipkins, R., Bolstad, R., Baker, R., Jones, A., Barker, M., Bell, B., et al. (2002) Curriculum, Learning and Effective Pedagogy: A Literature Review in Science Education. Wellington: Ministry of Education. Hipkins, R. and English, C. (2000) Building science concepts in primary school classrooms, in Ministry of Education (ed.) Exploring Issues in Science Education: Papers from a Research Seminar on Science Education in Primary Schools, 1999 (pp. 49–58). Wellington: Learning Media. Hodson, D. (1993) In search of a rationale for multicultural science education, Science Education, 77: 685–711. Hodson, D. (1998) Teaching and Learning Science: Towards a Personalised Approach. Buckingham: Open University Press. Hohepa, M., McNaughton, S. and Jenkins, K. (1996) Maori pedagogies and the roles of the individual, New Zealand Journal of Educational Studies, 31(1): 29–40. Hume, A. (1992) Earthquakes: A Teaching Resource. Hamilton: University of Waikato. Jegede, O. (1995) Collateral learning and the eco-cultural paradigm in science and mathematics education in Africa, Studies in Science Education, 25: 97–137. Jenkins, E. W. (2000) Constructivism in school science education: powerful tool or the most dangerous intellectual tendency, Science and Education, 9: 339–66. Jesson, J. (1990) Science Aotearoa: the politics of curriculum change. Unpublished MA thesis. University of Auckland, Auckland.
220
References
Jesson, J. (1995) Hot topic for discussion, New Zealand Herald, 24 June. Jones, A. (1982a) Physics at Work. Auckland: Science Centre. Jones, A. (1982b) School physics and physics in society – some aspects of the interrelationship. Unpublished MSc thesis, University of Waikato, Hamilton. Jones, A. (1988) Classroom perceptions of physics and the introduction of technological applications. Unpublished PhD thesis, University of Waikato, Hamilton. Jones, A. (1990) Teaching Physics from Technological Applications: A Teachers’ Guide. Hamilton: University of Waikato. Jones, A., Cowie, B. and Moreland, J. (2003) Enhancing formative interactions in science and technology: a synthesis of teacher student perspectives. Paper presented at the NARST Annual Conference, Philadelphia, March 23–26. Jones, A. and Osborne, R. (1985) Students’ perceptions of physics in society, Research and Development in Higher Education, 8: 348–54. Jones, A. T. and Kirk, C. M. (1989) Teaching technological applications in the physics classroom, Research in Science Education, 19: 164–73. Jones, A. T. and Kirk, C. M. (1990a) Gender differences in students’ interests in applications of school physics, Physics Education, 25(6): 308–13. Jones, A. T. and Kirk, C. M. (1990b) Introducing technological applications into the physics classroom. Help or hindrance to learning?, International Journal of Science Education, 12(5): 481–90. Keeves, J. and Alagumalai, S. (1998) Advances in measurement in science education, in B. J. Fraser and K. Tobin (eds) International Handbook of Science Education. Great Britain: Kluwer Academic Publishers. Kelly, G. (1969) Ontological acceleration, in B. Maher (ed.) Clinical Psychology and Personality: The Selected Papers of George Kelly ( pp. 15–24). New York, NY: Wiley. Kelly, G. A. (1955) The Psychology of Personal Constructs. New York, NY: Norton. Kent, L. (1996) How shall we know them? Comparison of Maori student responses for written and oral assessment tasks. Unpublished MEd thesis, University of Waikato, Hamilton. King, F. (1991) From the Weald to Wellington – a history of six pauper families who emigrated from Staplehurst in Kent to Port Nicholson in Wellington, arriving in April, 1840. Unpublished Diploma in Local History dissertation. University of Kent, Canterbury. King, M. (2003) The History of New Zealand. Auckland: Penguin. Kirkwood, V. (1988) Energy for a change: teaching and learning about the energy concept in junior secondary school science classrooms in New Zealand. Unpublished DPhil thesis, University of Waikato, Hamilton. Kirkwood, V. (1989) Energy For a Change: Teachers’ Guide. Hamilton: University of Waikato. Kirkwood, V., Bearlin, M. and Hardy, T. (1989) New approaches to the inservice education in science and technology for primary and early childhood teachers (or Mum is not so dumb after all!), Research in Science Education, 19: 174 –86. Kirkwood, V. and Carr, M. (1988a) Final Report: Learning in Science Project (Energy). Hamilton: University of Waikato. Kirkwood, V. and Carr, M. (1988b) A most valuable growing exercise, Research in Science Education, 19: 95–103. Kirkwood, V. and Carr, M. (1989) Research on Teaching and Learning Energy: Some Selected Papers of the Learning in Science (Energy) Project. Hamilton: University of Waikato.
References
221
Kirkwood, V., Carr, M., Bell, B., McChesney, J., Osborne, R. and Symington, D. (1985) LISP (Energy): The Strategy and Some Preliminary Findings. Working Paper 302, Learning in Science Project (Energy). Hamilton: University of Waikato. Kirkwood, V., Carr, M. and Gibb, S. (1987) Energy in a New Zealand Secondary School Junior Science Classroom: Case Study A. Working Paper 304. Learning in Science Project (Energy). Hamilton: University of Waikato. Kirkwood, V., Carr, M., Gibb, S., Shaw, J. and Stonyer, J. (1988) Energy in a New Zealand Secondary School Junior Science Classroom: Action Research of a Trial Unit. Working Paper 309, Learning in Science Project (Energy). Hamilton: University of Waikato. Kirkwood, V., Carr, M., Gibb, S. and Stonyer, J. (1988) Teachers’ Interaction with LISP (Energy). Working Paper 310, Learning in Science Project (Energy). Hamilton: University of Waikato. Kirkwood, V., Carr, M. and Newman, B. (1988) Energy in a New Zealand Secondary School Junior Science Classroom: A Trial Unit. Working Paper 308, Learning in Science Project (Energy). Hamilton: University of Waikato. Kirkwood, V., Carr, M. and Shaw, J. (1987) Energy in a New Zealand Secondary School Junior Science Classroom: Case Study C. Working Paper 306, Learning in Science Project (Energy). Hamilton: University of Waikato. Kirkwood, V., Carr, M. and Stonyer, J. (1987) Energy in a New Zealand Secondary School Junior Science Classroom: Case Study B. Working Paper 305, Learning in Science Project (Energy). Hamilton: University of Waikato. Kubli, F. (1979) Piaget’s cognitive psychology and its consequences for the teaching of science, European Journal of Science Education, 1(1): 5–20. Lal, K. (1991) Teachers’ views on internal assessment of practical work in sixth form chemistry. Unpublished MSc, University of Waikato. Lambert, J. and Tasker, R. (1982) Towards Solutions: The Work of the Science Action-research Group. Working Papers 46, 47, 48, 49. Learning in Science Project. Hamilton: University of Waikato. Lather, P. (1991) Getting Smart: Feminist Research and Pedagogy With/in the Postmodern. New York, NY: Routledge. Lave, J. and Wenger, E. (1991) Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press. Leach, J., Ametler, J., Hind, A., Lewis, J. and Scott, P. (2003) Evidence-informed approaches to teaching science at junior high school level: outcomes in terms of student learning. Paper presented at the annual meeting of the National Association for Research in Science Teaching, March, Philadelphia. Leach, J. and Moon, B. (1999a) Recreating pedagogy, in J. Leach and B. Moon (eds), Learners and Pedagogy. London: Paul Chapman Publishing. Leach, J. and Moon, B. (eds) (1999b) Learners and Pedagogy. London: Paul Chapman Publishing. Leach, J. and Scott, P. (2003) Individual and sociocultural views of learning in science education, Science and Education, 12: 91–113. Leach, J. and Scott, P. (in press) Designing and evaluating science teaching sequences: an approach drawing on the concept of learning demand and a social constructivist perspective on learning, Studies in Science Education. Lee, H. and Lee, G. (2001) The National Certificate of Educational Achievement (NCEA): ‘Fragile – handle with care’, New Zealand Annual Review of Education, 10: 5–38.
222
References
Leithwood, K. (1992) The principal’s role in teacher development, in M. Fullan and A. Hargreaves (eds) Teacher Development and Educational Change (pp. 86–103). London: Falmer Press. Lemke, J. (1990) Talking Science: Language, Learning and Values. Norwood, NJ: Ablex Publishing Corporation. Lemke, J. (2001) Articulating communities: sociocultural perspectives on science education, Journal of Research in Science Teaching, 38(3): 296–316. Linn, R. L., Baker, E. and Dunbar, S. (1991) Complex, peformance-based assessment: expectations and validity criteria, Education Researcher, 20: 15–21. Loo, S. P. (2001) Islam, science and science education, Studies in Science Education, 36: 45. Lorsbach, A., Tobin, K., Briscoe, C. and LaMaster, S. (1992) An interpretation of assessment methods in middle school science, International Journal of Science Education, 14: 305–17. Loughran, J. J., Hamilton, M. L., LaBoskey, V. K. and Russell, T. L. (in press) The International Handbook of Self-Study of Teaching and Teacher Education Practices, Volumes 1 and 2. Dordrecht: Kluwer Academic Publishers. Loughran, J. J. and Russell, T. L. (eds) (2002) Improving Teacher Education Practices Through Self-study. London: Routledge-Falmer. Lovegrove, M. N. (1966) The scholastic achievement of European and Maori children, New Zealand Journal of Educational Studies, 1(1): 16–39. Loveless, M. and Barker, M. (2000) ‘Those pages we just turn over . . .’: the nature of science in Science in the New Zealand Curriculum, New Zealand Science Teacher, 93: 28–32. Luke, C. and Gore, J. (1992) Feminisms and Critical Pedagogy. New York, NY: Routledge. Manu’atu, L. (2000) Katoanga Faiva: a pedagogical site for Tongan students, Educational Philosophy and Theory, 32(1): 73–80. Maori Language Commission (2002) Press release: Maori most widely spoken second language. Wellington: Te Taura Whiri Reo Maori (Maori Language Commission). Matthews, M. (1992) Constructivism and empiricism: an incomplete divorce, Research in Science Education, 22: 299–307. Matthews, M. (1993) Curriculum reform degrades sciences, New Zealand Herald, 26 August. Matthews, M. (1994) Science Teaching: The Role of the History and Philosophy of Science. New York, NY: Routledge. Matthews, M. (1995) Challenging NZ Science Education. Palmerston North, NZ: The Dunmore Press. Matthews, M. (ed.) (1998) Constructivism in Science Education: A Philosophical Examination. Dordrecht: Kluwer Academic Publishers. May, S. (2000) Uncommon languages: the challenges and possibilities of minority language rights, Journal of Multilingual and Multicultural Development, 21(5): 366–85. Mays, W. (1982) Piaget’s sociological theory, in S. Modgill and C. Modgill (eds) Jean Piaget: Consensus and Controversy (pp. 31–50). London: Holt, Rinehart and Winston. McCan, D. (2001) Whatiwhatihoe: The Waikato Raupatu Claim. Wellington: Huia Publishers. McComish, J. (1994) What has constructivism got to do with feminism? Paper presented at the Feminism/Science Education Symposium, University of Waikato, Hamilton. McCraw, J. (1990) Maori legends as an aid to teaching earth science, New Zealand Science Teacher, 65: 45– 7. McGee, C. (1997) Teachers and Curriculum Decision-making. Palmerston North, NZ: The Dunmore Press.
References
223
McGee, C., Jones, A., Cowie, B., Hill, M., Miller, T., Harlow, A. and Mackenzie, K. (2003) Teachers’ Experiences in Implementing Science in the New Zealand National Curriculum. Hamilton: University of Waikato. McKinley, E. (1992a) Research issues in Maori science education, SAMEpapers, 92: 1– 21. McKinley, E. (1992b) Teaching primary science in Maori. Unpublished directed research paper. University of Waikato, Hamilton. McKinley, E. (1995a) Maori and science education, Access, 13(2): 122–5. McKinley, E. (1995b) A power/knowledge nexus: writing a science curriculum in Maori. Unpublished MEd, University of Waikato. McKinley, E. (1997) Maori and science education, in B. Bell and R. Baker (eds) Developing the Science Curriculum in Aotearoa New Zealand (pp. 213–26). Auckland: Addison Wesley Longman New Zealand. McKinley, E. (1998) Science curricula and cultural diversity: are we doing enough for the aspirations of Maori?, in D. Hodson (ed.) Science and Technology Education and Ethnicity: An Aotearoa/New Zealand Perspective. Wellington: Royal Society of New Zealand. McKinley, E. (2000) Maori science education: the urgent need for research, in Ministry of Education (ed.) Exploring Issues in Science Education. Wellington: Learning Media. McKinley, E. (2001) Cultural diversity: masking power with innocence, Science Education, 85(5): 74–6. McKinley, E. (2003) Brown bodies, white coats: postcolonialism, Maori women and science. Unpublished PhD thesis, University of Waikato, Hamilton. McKinley, E. and McPherson Waiti, P. (1993) Editorial, SAMEpapers, 93: 1–4. McKinley, E., McPherson Waiti, P. and Bell, B. (1992) Language, culture and science education, International Journal of Science Education, 14(5): 579–95. McKinley, E. and Waiti, P. (1995) Te Tauki Maratautanga Putaiao: He tauira – the writing of a national science curriculum in Maori, SAMEpapers, 95: 75–94. McKinley, L. and Devine, N. (2001) The gap between rhetoric and practice: working paper on ‘Experiences of Maori Students in Classrooms. Paper presented at the ‘Disrupting preconceptions’ post-colonial conference, University of Queensland, Brisbane, Australia. McMillan, B. (1995) Making sense of the debate about the Science Curriculum document, SAMEpapers, 95: 95–113. McNaught, C. (1993) Which science? Which language? SAMEpapers, 93: 148–71. Messick, S. (1989) Validity, in R. L. Linn (ed.) Educational Measurement in Education. Washington, DC: American Council on Education and National Council on Measurement in Education. Messick, S. (1994) The interplay of evidence and consequences in the validation of performance assessment, Educational Researcher, 23(2): 13–23. Metge, J. (1980) Multi-culturalism: problem or goal? In National Advisory Committee on Maori Education (ed.) He Huarahi: Report of the National Advisory Committee on MAori Education. Wellington: Department of Education. Metge, J. (1984) He Tikanga MAori: Learning and Teaching. Wellington: Department of Education. Middleton, S. (1993) Educating Feminists: Life Histories and Pedagogies. New York, NY: Teachers’ College Press. Millar, R. (1989) Constructive criticisms, International Journal of Science Education, 11(5): 587–96.
224
References
Millar, R. and Driver, R. (1987) Beyond processes, Studies in Science Education, 14: 33–62. Ministry of Education (1990a) Draft Forms 1–5 Science Syllabus. Wellington: Learning Media. Ministry of Education (1990b) Learning in Science: A Professional Development Guide for Teachers of Science. Wellington: Learning Media. Ministry of Education (1990c) A Local Science Network: A Professional Guide for Teachers of Science. Wellington: Learning Media. Ministry of Education (1990d) Science Aotearoa: A Professional Development Guide for Teachers of Science. Unpublished page proofs. Wellington: Learning Media. Ministry of Education (1990e) Tomorrow’s Standards: The Report of the Ministerial Working Party on Assessment for Better Learning. Wellington: Learning Media. Ministry of Education (1991a) Girls and Science: A Professional Guide for Teachers of Science. Wellington: Learning Media. Ministry of Education (1991b) Learning Strategies and Teaching Approaches: A Professional Guide for Teachers of Science. Wellington: Learning Media. Ministry of Education (1991c) The Role of Practical Work in Science: A Professional Guide for Teachers of Science. Wellington: Learning Media. Ministry of Education (1991d) Science and Technology: A Professional Guide for Teachers of Science. Wellington: Learning Media. Ministry of Education (1992a) Born Too Soon. Wellington: Learning Media. Ministry of Education (1992b) Diving for Gold. Wellington: Learning Media. Ministry of Education (1992c) Hot Metal. Wellington: Learning Media. Ministry of Education (1992d) Near Miss. Wellington: Learning Media. Ministry of Education (1992e) Science in the National Curriculum: Draft. Wellington: Learning Media. Ministry of Education (1993a) The New Zealand Curriculum Framework. Wellington: Learning Media. Ministry of Education (1993b) Science in the New Zealand Curriculum. Wellington: Learning Media. Ministry of Education (1994a) Assessment: Policy and Practice. Wellington: Learning Media. Ministry of Education (1994b) Biology in the New Zealand Curriculum. Wellington: Learning Media. Ministry of Education (1995a) Developing Science Programmes. Wellington: Learning Media. Ministry of Education (1995b) Investigating Science. Wellington: Learning Media. Ministry of Education (2002) Curriculum Stocktake Report to Minister of Education. Wellington: Ministry of Education: http://www.minedu.govt.nz. Ministry of Education (2003a) MAori Medium Education at 1 July, 2003. Wellington: Ministry of Education. Ministry of Education (2003b) The New Zealand Curriculum Project. Wellington: Ministry of Education: http://www.tki.govt.nz/r/nzcurriculum. Ministry of Education (2003c) Teaching and Learning Research Initiative. Wellington: http:// www.tlri.org.nz. Mohamed, A. (1995) Learning physics in a technological context: a cross cultural comparative study. Unpublished PhD thesis, University of Waikato, Hamilton. Moli, S. (1993a) Some Samoan science teachers’ views of teaching and learning. SAMEpapers, 93: 84–103. Moli, S. (1993b) Towards a relevant science education for Western Samoa. Unpublished MEd thesis, University of Waikato, Hamilton.
References
225
Monk, G., Winslade, J., Crockett, K. and Epston, D. (eds) (1997) Narrative Therapy in Practice: the Archeology of Hope. San Francisco, CA: Jossey-Bass. Moreland, J. and Jones, A. (2000) Emerging assessment practices in an emergent curriculum: implications for technology, International Journal of Technology and Design Education, 10(3): 283–305. Moreland, J., Jones, J. and Northover, A. (2001) Enhancing teachers’ technological knowledge and assessment practices to enhance student learning in technology: a two year classroom study, Research in Science Education, 31(1): 155–76. Morgan, C. and Morris, G. (1999) Good Teaching and Learning: Pupils and Teachers Speak. Buckingham: Open University Press. Moss, P. A. (1992) Shifting conceptions of validity in educational measurement: implications of performance assessment, Review of Educational Research, 62(3): 229–58. Moyle, R. (1980) Weather. Working Paper 21, Learning in Science Project. Hamilton: University of Waikato. Nadeau, R. and Desautels, J. (1984) Epistemology and the Teaching of Science. Ottawa: Science Council of Canada. National Advisory Committee on Maori Education (1980) He Huarahi. Wellington: Department of Education. National Research Council (1999) The Assessment of Science Meets the Science of Assessment. Washington, DC: National Academy Press. Newmann, D., Griffin, P. and Cole, M. (1989) The Construction Zone: Working for Cognitive Change in School. Cambridge: Cambridge University Press. Niaz, M., Abd-El-Khalick, F., Benarroch, A., Cardellini, L., Laburú, C. E., Marín, N., Montes, L. A., Nola, R., Orlik, Y., Scharmann, L. C., Tsai, C.-C. and Georgios Tsaparlis, G. (2003) Constructivism: defense or a continual critical appraisal – a response to Gil-Perez et al., Science and Education, 12: 787–97. Nola, R. (1995) Against constructivism in the philosophy of science education, Access, 13(2): 21–36. Northfield, J. and Symington, D. (1991) Learning in Science Viewed as a Personal Construction: An Australian Perspective. SMEC Centre, Curtin University. Novak, J. (1977) A Theory of Education. Ithaca, NY: Cornell University Press. Novak, J. (1978) An alternative to Piagetian psychology for science and mathematics education, Studies in Science Education, 5: 1–30. Nussbaum, J. (1989) Classroom conceptual change: philosophical perspectives, International Journal of Science Education, 11(special issue): 530–40. Nussbaum, J. and Novick, S. (1981) Brainstorming in the classroom to invent a model: a case study, School Science Review, 62(221): 77–8. Nuthall, G. (1997) Understanding student thinking and learning in the classroom, in B. Biddle, T. Good and I. Goodson (eds) The International Handbook of Teachers and Teaching. Dordrecht: Kluwer Academic Publishers. Nuthall, G. and Alton-Lee, A. (1993) Predicting learning from student experience of teaching: a theory of student knowledge construction in classrooms, American Educational Research Journal, 30: 799 –840. Ohia, M. (1990) The unresolved conflict and debate: an overview of bilingual education in New Zealand secondary schools, SAMEpapers, 90: 111–32. O’Loughlin, M. (1992) Rethinking science education: beyond Piagetian constructivism towards a socio-cultural model of teaching and learning, Journal of Research in Science Teaching, 29(8): 791– 820.
226
References
Olssen, M. (1991) Producing the truth about people: science and the cult of the individual in educational psychology, in J. Morss and T. Linzey (eds) Growing Up: the Politics of Human Learning. Auckland, NZ: Longman Paul. Olssen, M. (1995a) The epistemology of constructivism, Access, 13(2): 49–63. Olssen, M. (1995b) Wittgenstein and Foucault: the limits and possibilities of constructivism, Access, 13(2): 71–8. Osborne, J. (1996) Beyond constructivism, Science Education, 80(1): 53–82. Osborne, J. and Ratcliffe, M. (2002) Developing effective methods of assessing ideas and evidence, School Science Review, 83(305): 113–23. Osborne, R. (1976) Using student attitudes to modify instruction in physics, Journal of Research in Science Teaching, 13(6): 525–31. Osborne, R. (1980a) Force. Working Paper 16, Learning in Science Project. Hamilton: University of Waikato. Osborne, R. (1980b) Some aspects of students’ view of the world, Research in Science Education, 10: 11–18. Osborne, R. (1981) Children’s ideas about electric current, NZ Science Teacher, 29: 12–19. Osborne, R. (1982a) Conceptual change for pupils and teachers, Research in Science Education, 12: 25–41. Osborne, R. (1982b) Towards a Framework for the Learning in Science Project. Working Paper 101, Learning in Science Project (Primary). Hamilton: University of Waikato. Osborne, R. (1983) Towards modifying children’s ideas about electric current, Research in Science and Technology Education, 1(1): 73–82. Osborne, R. (1984) Science in the Primary School. Working Paper 131, Learning in Science Project (Primary). Hamilton: University of Waikato. Osborne, R. (1985) Children’s own concepts, in W. Harlen (ed.) Primary Science: Taking the Plunge. London: Heinemann. Osborne, R., Bell, B. and Gilbert, J. (1983) Science teaching and children’s views of the world, European Journal of Science Education, 5(1): 1–14. Osborne, R. and Biddulph, F. (1985a) Children’s Ideas: Selected Working Papers from the Learning in Science Project (Primary). Hamilton: University of Waikato. Osborne, R. and Biddulph, F. (1985b) Final Report: Learning in Science Project (Primary). Hamilton: University of Waikato. Osborne, R., Biddulph, F., Freyberg, P. and Symington, D. (1982) Confronting the Problems of Primary Science. Working Paper 110, Learning in Science Project (Primary). Hamilton: University of Waikato. Osborne, R. and Cosgrove, M. (1983) Children’s conceptions of the changes of states of water, Journal of Research in Science Teaching, 20(9): 825–38. Osborne, R. and Freyberg, P. (1985) Learning in Science: The Implications of Children’s Science. Auckland: Heinemann. Osborne, R., Freyberg, P. and Tasker, R. (1980) Looking at Problems: The Working Papers of the In-depth Phase. Learning in Science Project. Hamilton: University of Waikato. Osborne, R., Freyberg, P. and Tasker, R. (1982) Towards Solutions: The Working Papers of the Action-research Phase. Learning in Science Project (F1–4). Hamilton: University of Waikato. Osborne, R., Freyberg, P., Tasker, R. and Stead, K. (1981) Description, analysis and action: three phases of a research project, Research in Science Education, 11: 52–8. Osborne, R. and Gilbert, J. (1979) Investigating student understanding of basic concepts using an interview-about-instances approach, Research in Science Education, 9: 85–93.
References
227
Osborne, R. and Gilbert, J. (1980a) A method for the investigation of concept understanding in science, European Journal of Science Education, 2(3): 311–21. Osborne, R. and Gilbert, J. (1980b) A technique for exploring students’ views of the world, Physics Education, 50(6): 376–9. Osborne, R. and Schollum, B. (1981) Towards Solutions: The Work of the Physics Action-research Group. Working Papers no. 32, 33, 34. Learning in Science Project. Hamilton: University of Waikato. Osborne, R., Schollum, B. and Russell, S. (1982) Towards Solution: The Work of the Chemistry Action-research Group. Working Papers 35, 36, 37, 38. Learning in Science Project. Hamilton: University of Waikato. Osborne, R. and Tasker, R. (1980) A Visit to a Classroom: No. 2. Working Paper 39, Learning in Science Project. Hamilton: University of Waikato. Osborne, R., Tasker, R. and Stead, K. (1979) Visit to a Classroom. Working Paper 13, Learning in Science Project. Hamilton: University of Waikato. Osborne, R. and Wittrock, M. (1983) Learning science: a generative process, Science Education, 67(4): 489–508. Osborne, R. and Wittrock, M. (1985) The generative learning model and its implications for science education, Studies in Science Education, 12: 59–87. Oxenham, L. (1995) Introducing science ideas: an investigation into methods used by teachers to introduce science ideas in student-centred teaching. Unpublished MSc thesis, University of Waikato, Hamilton. Pea, R. (1993) Distributed intelligence and designs for education, in G. Salomon (ed.) Distributed Cognitions: Psychological and Educational Considerations. New York, NY: Cambridge University Press. Pearson, J. and Bell, B. (1993) The Teacher Development That Occurred. A Report of the Learning in Science Project (Teacher Development). Hamilton: University of Waikato. Pere, R. (1982) Ako: Concepts and Learning in the MAori Tradition. Hamilton: University of Waikato. Perkins, D. (1993) Person-plus; a distributed view of thinking and learning, in G. Salomon (ed.) Distributed Cognitions: Psychological and Educational Considerations. New York, NY: Cambridge University Press. Perkins, D., Jay, E. and Tishman, S. (1993) Beyond abilities: a dispositional theory of thinking, The Merill-Palmer Quarterly, 39(1): 1–21. Perkins, D. and Salomon, G. (1989) Are cognitive skills context bound?, Educational Researcher, 18(1): 16–25. Pfundt, H. and Duit, R. (1994) Bibliography: Students’ Alternative Framework and Science Education. Kiel: IPN. Phillips, D. (1998) Recent developments in national assessment policy in education, New Zealand Annual Review of Education, 8: 147–67. Piaget, J. (1970) Piaget’s theory, in P. Mussen (ed.) Carmichael’s Manual of Child Psychology (pp. 61–84). New York, NY: Wiley. Pinar, W. F. (ed.) (1998) Queer Theory in Education. London: Lawrence Erlbaum Associates. Pinar, W. F., Reynolds, W. M., Slattery, P. and Taubman, P. (1995) Understanding Curriculum: An Introduction to the Study of Historical and Contemporary Curriculum. New York, NY: Peter Lang. Pintrich, P., Marx, R. and Boyle, R. (1993) Beyond cold conceptual change: the role of motivational beliefs and classroom contextual factors in the process of conceptual change, Review of Educational Research, 63(2): 167–99.
228
References
Pollard, A., Triggs, P., Broadfoot, P., McNess, E. and Osborn, M. (2000) What Pupils Say: Changing Policy and Practice in Primary Education. London: Continuum. Pope, M. and Gilbert, J. (1983) Personal knowledge and the construction of knowledge in science, Science Education, 67(2): 193–203. Pope, M. and Keen, T. (1981) Personal Construct Psychology and Education. London: Academic Press. Portelli, J. (1993) Exposing the hidden curriculum, Journal of Curriculum Studies, 25(4): 343–58. Porteous, D. (1997) Choosing physics and physics learning contexts. Unpublished MSc thesis, University of Waikato, Hamilton. Poschalk, K. (1986) Energy in the Primary School. Working Paper 303, Learning in Science Project (Energy). Hamilton: University of Waikato. Posner, G., Strike, K., Hewson, P. and Gertzog, W. (1982) Accommodation of a scientific conception: toward a theory of conceptual change, Science Education, 66(2): 211–27. Power, C. (1972) ASEP as a stimulus for research – a progress report of a study of teaching strategies used with ASEP materials, Research in Science Education, 2: 13–20. Pryor, J. and Torrance, H. (2000) Questioning the three bears: the social construction of classroom assessment, in A. Filer (ed.) Assessment: Social Practice and Social Product. London: RoutledgeFalmer. Raghavan, N. (1988) Peer interaction and conceptual change: students’ views of ‘living’. Unpublished MEd thesis, University of Waikato, Hamilton. Raven, R. J. (1967–68) The development of the concept of momentum in primary school children, Journal of Research in Science Teaching, 5: 216–23. Rawaikela Dakuidreketi, Mesake (1995) Factors contributing to academic achievement of first year native-Fijian students in science courses at the University of the South Pacific. Unpublished MSc thesis, University of Waikato, Hamilton. Reay, D. and Wiliam, D. (1999) ‘I’ll be a nothing’: structure, agency and the construction of identity through assessment [1], British Educational Research Journal, 25(3): 343–54. Resnick, L. (1991) Shared cognition: thinking as social practice, in L. Resnick, J. Levine and S. Teasley (eds) Perspectives on Socially Shared Cognition. Washington: American Psychological Association. Rikihana, H. T. (no date) Science ( Vols 1 and 2). Auckland: Auckland College of Education. Roberts, M. (1993) Scientific knowledge and cultural traditions. Part II: A Pakeha view of the kiore rat in New Zealand, SAMEpapers, 93: 23–45. Rodrigues, S. (1993a) Oxidation and Reduction: Some Six Form Units. Hamilton: University of Waikato. Rodrigues, S. (1993b) The role and effect of context in learning Form Six oxidation and reduction concepts by female students. Unpublished DPhil thesis, University of Waikato, Hamilton. Rodrigues, S. and Bell, B. (1995) Chemically speaking: a description of student-teacher talk during chemistry lessons using and building on students’ experiences, International Journal of Science Education, 17(6): 797–809. Rogoff, B. (1993) Children’s guided participation and participatory appropriation of sociocultural activity, in R. Wozniak and K. Fischer (eds) Development in Context: Acting and Thinking in Specific Environments. Hillsdale, NJ: Lawrence Erlbaum Associates. Rogoff, B. (1995) Observing sociocultural activity on three planes, in J. Wertsch, P. Del Río and A. Alvarez (eds) Sociocultural Studies of the Mind. Cambridge: Cambridge University Press.
References
229
Roth, W.-M. (1995) Authentic School Science: Knowing and Learning in Open-inquiry Science Laboratories. Dordrecht: Kluwer Academic Publishers. Roth, W.-M. (1998) Designing Communities. Dordrecht: Kluwer Academic Publishers. Roth, W.-M. and Roychoudhury, A. (1994) Physics students’ epistemologies and views of knowing and learning, Journal of Research in Science Teaching, 31(1): 5–30. Rowell, J. and Dawson, C. (1984) Equilibration, conflict and instruction: a new class orientated perspective, European Journal of Science Education, 7(4): 331–44. Rutherford, M. (1993) The RADMASE Centre: some linguistic aspects for material production for pre and in-service teacher education. SAMEpapers, 93: 138–47. Sadler, R. (1989) Formative assessment and the design of instructional systems, Instructional Science, 18(2): 119–44. Saeed, S. (1997) Practical work in school science: a qualitative study of teachers’ perceptions. Unpublished MSc thesis, University of Waikato, Hamilton. Salomon, G. (1993a) No distribution without individuals’ cognition, in G. Salomon (ed.) Distributed Cognitions: Psychological and Educational Considerations. New York, NY: Cambridge University Press. Salomon, G. (ed.) (1993b) Distributed Cognitions: Psychological and Educational Considerations. New York, NY: Cambridge University Press. Salomon, G. and Perkins, D. (1998) Individual and social aspects of learning, Review of Research in Education, 23: 1–24. Sanjek, R. (ed.) (1990) Fieldnotes: The Makings of Anthropology. Ithaca, NY: Cornell University Press. Sarf, A. (1998) On two metaphors for learning and the dangers of choosing just one, Educational Researcher, 27: 4–13. Scheurich, J. and Young, M. (1997) Coloring epistemologies: are our research epistemologies racially biased?, Educational Researcher, 26(4): 4–16. Schollum, B. (1982a) Chemical change, NZ Science Teacher, 33: 5–9. Schollum, B. (1982b) Chemical Change. Working Paper 27, Learning in Science Project. Hamilton: University of Waikato. Schollum, B. (1986) An analysis of a common activity: growing crystals. Unpublished MEd thesis, University of Waikato, Hamilton. Schollum, B. (no date) Growing Crystals: A Resource Unit for Teachers. Hamilton: University of Waikato. Schutz, A. and Luckmann, T. (1973) Structures of the Life World. London: Heinemann. Scott, P. (1998) Teacher talk and meaning making in science classrooms: a Vygotskian analysis and review, Studies in Science Education, 32: 45–80. Scott, P. (1999) An analysis of science classroom talk in terms of the authoritative and dialogic nature of the discourse. Paper presented at the NARST Annual Meeting, Boston, USA, April. Shaw, J. (1997) Threats to the validity of science performance assessments for English language learners, Journal of Research in Science Teaching, 34(7): 721–43. Shayer, M. and Adey, P. (1981) Towards a Science of Science Teaching. London: Heinemann. Shipstone, D. (1984) A study of children’s understanding of electricity in simple DC circuits, European Journal of Science Education, 6(2): 185–98. Shulman, L. (1987) Knowledge and teaching: foundations of the new reforms, Harvard Educational Review, 57: 1–22. Silvester, J. (1988a) The Interviews. Working Paper 405, Learning in Science Project (Teaching). Hamilton: University of Waikato.
230
References
Silvester, J. (1988b) The Literature Review. Working Paper 404, Learning in Science Project (Teaching). Hamilton: University of Waikato. Singh, A. (1993) Problem solving in physics: an interactive computer assisted environment. Unpublished PhD thesis, University of Waikato, Hamilton. Singh, A. and Carr, M. (1992) Learning theories and environments; a student-initiated intelligent computer assisted environment, Research in Science Education, 22: 367–76. Smith, G. (1986) Nga Kete Wanaga Readers in MAori Education, vols 1, 2, 3. Auckland: Auckland College of Education. Smith, G. H. (1990) Taha Maori: Pakeha capture, in J. Codd, R. Harker and R. Nash (eds) Political Issues in New Zealand Education, 2nd edn. (pp. 183–97). Palmerston North: The Dunmore Press. Smith, G. H. (1997) The development of Kaupapa Maori: theory and praxis. Unpublished PhD thesis, International Research Institute for Maori and Indigenous Education, University of Auckland, Auckland. Solano-Flores, G. and Nelson-Barber, S. (2001) On the cultural validity of science assessments, Journal of Research in Science Teaching, 38(5): 553–73. Solomon, J. (1983) Learning about energy: how pupils think in two domains, European Journal of Science Education, 5(1): 49–59. Solomon, J. (1987) Social influences on the construction of pupils’ understanding of science, Studies in Science Education, 14: 63–82. Solomon, J. (1994) The rise and fall of constructivism, Studies in Science Education, 23: 1–19. Spolsky, B. (1987) Report of MAori–English Bilingual Education. Wellington: Department of Education. Standrill, B. (1983) Students’ concepts of chemical stability. Unpublished MSc thesis, University of Waikato, Hamilton. Stanley, W. and Brickhouse, N. (2001) Teaching sciences: the multicultural question revisited, Science Education, 85: 35–49. Statistics New Zealand (2001) Health of the MAori Language. Wellington: Statistics New Zealand. Stavy, R. and Berkovitz, B. (1980) Cognitive conflict as a basis for teaching quantitative aspects of the concept of temperature, Science Education, 64(5): 679–92. Stead (now Bell), B. (1980a) The description and modification of some students’ biological concepts. Unpublished MEd thesis, University of Waikato, Hamilton. Stead (now Bell), B. (1980b) Energy. Working Paper 17, Learning in Science Project. Hamilton: University of Waikato. Stead (now Bell), B. and Osborne, R. (1979) Some students’ concepts of light. Paper presented at the First Conference of the New Zealand Association for Research in Education, December, Wellington, NZ. Stead (now Bell), B. and Osborne, R. (1980) Exploring students’ concepts of light, Australian Science Teachers’ Journal, 26(3): 84–90. Stead, K. (1982) Outlook on Science. Working Paper 50, Learning in Science Project. Hamilton: University of Waikato. Stead, K. (1983) Insights into students’ outlooks on science with personal constructs, Research in Science Education, 13: 163–76. Stead, K. (1984) An exploration of different outlooks on science: towards an understanding of the under-representation of girls and of Maori and Pacific Island students in science. Unpublished DPhil thesis, University of Waikato, Hamilton. Stead, K. and Osborne, R. (1981a) Friction. Working Paper 19, Learning in Science Project. Hamilton: University of Waikato.
References
231
Stead, K. and Osborne, R. (1981b) Gravity. Working Paper 20, Learning in Science Project. Hamilton: University of Waikato. Stewart, G. (in progress) Doctoral thesis research. University of Waikato, Hamilton. Stoddart, T., Abrams, R., Gasper, E. and Canaday, D. (2000) Concept maps as assessment in science inquiry learning – a report of methodology, International Journal of Science Education, 22(12): 1221–46. Strachan, J. (2002) Assessment in change: some reflections on the local and international background to the national certificate of Educational Achievement, New Zealand Annual Review of Education, 11: 245–62. Suchting, W. (1992) Constructivism deconstructed, Science and Education, 1: 223–54. Symington, D., Biddulph, F., Happs, J. and Osborne, R. (1983) Primary School Pupils’ Ideas about Rocks. Working Paper 107, Learning in Science Project (Primary). Hamilton: University of Waikato. Symington, D. and Osborne, R. (1984) Towards Professional Development in Science Education for the Primary School Teacher. Working Paper 119, Learning in Science Project (Primary). Hamilton: University of Waikato. Symington, D., Osborne, R., Freyberg, P. and White, R. (1982) A Perspective on Primary Science. Working Paper 104, Learning in Science Project (Primary). Hamilton: University of Waikato. Taba, H. (1962) Curriculum Development: Theory and Practice. New York, NY: Harcourt Brace and World. Tasker, R. (1980) Some aspects of the students’ view of doing science, Research in Science Education, 10: 19–22. Tasker, R. (1981) Children’s views and classroom experiences, Australian Science Teachers’ Journal, 27(3): 33–7. Tasker, R. (1982) Children’s Views and Classroom Experiences. Working Paper 43, Learning in Science Project. Hamilton: University of Waikato. Tasker, R. and Freyberg, P. (1985) Facing mismatches in the classroom, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implications of Children’s Science (pp. 66–80). Auckland: Heinemann. Tasker, R. and Osborne, R. (1982a) Portraying Pupils’ Classroom Experiences. Working paper 42, Learning in Science Project. Hamilton: University of Waikato. Tasker, R. and Osborne, R. (1982b) A Visit to a Classroom: No. 3. Working Paper 40. Learning in Science Project. Hamilton: University of Waikato. Tasker, R. and Osborne, R. (1982c) A Visit to a Classroom: No. 4. Working Paper 41. Learning in Science Project. Hamilton: University of Waikato. Tasker, R. and Osborne, R. (1983) Portraying pupils’ classroom experiences, Research in Science and Technology Education, 1(2): 133–44. Tasker, R. and Osborne, R. (1985) Science teaching and science learning, in R. Osborne and P. Freyberg (eds) Learning in Science: The Implications of Children’s Science. Auckland: Heinemann. Taylor, I. (2000) Promoting mental model-building in astronomy education. Unpublished PhD thesis, University of Waikato, Hamilton. Taylor, I., Barker, M. and Jones, A. (2003) Promoting mental model-building in astronomy education, International Journal of Science Education, 25(10): 1205–26. Taylor, P. (1991) Collaborating to reconstruct teaching: the influence of researcher beliefs. Paper presented at the annual meeting of the American Educational Research Association, Chicago, 3–7 April.
232
References
Te Tahuhu O te Matauranga (1996) Te TauAkI Marautanga PUtaiao. Whanganui a Tara, NZ: Learning Media. Te Tahuhu O te Matauranga (1997) Hangarau: Whakatene, te rea. Whanganui a Tara, NZ: Huia Publishers. Te Tahuhu O te Matauranga (2000) He Taura PUtaiao. Whanganui a Tara: Aotearoa. Thaman, K. H. (1995) Concepts of learning, knowledge and wisdom in Tonga and their relevance to modern education, Prospects, 15(4): 723–33. Thaman, K. H. (1997) Kakala: towards a Pacific concept of teaching and learning. Paper presented at the keynote address to the Australasian College of Education National Conference, Cairns, 28 September–1 October. Tishman, S., Jay, E. and Perkins, D. (1993) Teaching thinking dispositions: from transmission to enculturation, Theory into Practice, 32(3): 147–53. Tishman, S., Perkins, D. and Jay, E. (1995) The Thinking Classroom. Boston, MA: Allyn and Bacon. Tittle, C. (1994) Toward an educational psychology of assessment for teaching and learning: theories, contexts and validation arguments, Educational Psychologist, 29(3): 149–62. Tobin, K. (1990) Social constructivist perspectives on the reform of science education, The Australian Science Teachers’ Journal, 36(4): 29–35. Tobin, K. (1993) Referents for making sense of science teaching, International Journal of Science Education, 15(3): 241–54. Torrance, H. (1993) Formative assessment: some theoretical problems and empirical questions, Cambridge Journal of Education, 23(3): 333–43. Torrance, H. (2000) Post-modernism and educational assessment, in A. Filer (ed.) Assessment: Social Practice and Social Product. London: RoutledgeFalmer. Torrance, H. and Pryor, J. (1995) Investigating teacher assessment in infant classrooms: methodological problems and emerging issues, Assessment in Education, 2(3): 305–20. Torrance, H. and Pryor, J. (1998) Investigating Formative Assessment: Teaching, Learning and Assessment in the Classroom. Buckingham: Open University Press. Torrance, H. and Pryor, J. (2001) Developing formative assessment in the classroom: using action research to explore and modify theory, British Educational Research Journal, 27(5): 615–31. Torrie, I. (1989) Developing achievement-based assessment using grade-related criteria, Research in Science Education, 19: 286–90. Treagust, D. F., Jacobowitz, R., Gallagher, J. L. and Parker, J. (2001) Using assessment as a guide in teaching for understanding: a case study of a middle school science class learning about sound, Science Education, 85(2): 137–57. Trinick, T. (no date) Te Mahi Taiao: Te Pukapuka Tuatahi, Tuarua. Auckland: Nga Tapuwae College. Tyler, R. (1949) Basic Principles of Curriculum and Instruction. Chicago, IL: University of Chicago Press. Ume, J. (1996) Science teachers’ views of the Papua New Guinea (PNG) Science Curriculum and teacher development. Unpublished MSc thesis, University of Waikato, Hamilton. von Glaserfeld, E. (1991) Knowing without metaphysics: aspects of the radical constructivist position, in F. Steier (ed.) Research and Reflexivity (pp. 12–29). London: Sage Publications. Vygotsky, L. (1978) Mind in Society: The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press.
References
233
Waiti, P. (1994) Te Wananga Putaiao: a case study of an intervention. Unpublished directed study. University of Waikato, Hamilton. Walker, R. (1985) Cultural domination of Taha Maori: the potential for radical transformation, in J. Codd, R. Harker and R. Nash (eds) Political Issues in New Zealand Education. Palmerston North: The Dunmore Press. Walker, R. (1991) Liberating MAori from Educational Subjection. Monograph 6. Auckland: Research Unit for Maori Education, University of Auckland. Walkerdine, V. (1984) Developmental psychology and the child-centred pedagogy: the insertion of Piaget into early childhood education, in J. Henriques, W. Holloway, C. Urwin, C. Venn and V. Walkerdine, Changing the Subject. London: Methuen. Watts, M. and Bentley, D. (1987) Constructivism in the classroom: enabling conceptual development by words and deeds, British Educational Research Journal, 13(2): 121–35. Welch, W. (1985) Learning about Energy: A Review of the Literature. Working Paper 301, Learning in Science Project (Energy). Hamilton: University of Waikato. Wertsch, J. (1991) Voices of the Mind. Cambridge, MA: Harvard University Press. Wertsch, J., Del Río, P. and Alvarez, A. (1995) Sociocultural studies: history, action and mediation, in J. Wertsch, P. Del Río and A. Alvarez (eds) Sociocultural Studies of Mind. Cambridge: Cambridge University Press. West, L. and Pines, A. (1983) ‘How rational is rationality?’, Science Education, 67(1): 37– 9. Wetherell, M. and Potter, J. (1992) Mapping the Language of Racism: Discourse and the Legitimation of Exploitation. Hemel Hempstead: Harvester Wheatsheaf. Wheatley, G. (1991) Constructivist perspectives on science and mathematics learning, Science Education, 75(1): 9–21. White, R. (1974) The validation of learning hierarchy, American Educational Research Journal, 11(2): 121–36. White, R. (1988) Learning Science. Cambridge, MA: Basil Blackwell. White, R. (1991) An overview of the Australasian perspective, in J. Northfield and D. Symington (eds) Learning in Science Viewed as Personal Construction: An Australasian Perspective (pp. 1–15). Perth: SMEC Centre, Curtin University of Technology. White, R. and Gunstone, R. (1989) Metalearning and conceptual change, International Journal of Science Education, 11, special issue: 577–86. White, R. and Gunstone, R. (1992) Probing Understanding. London: Falmer Press. Wigglesworth, P. Z. (1999) Aspects of the classroom environment and their perceived impact on science learning. Unpublished MPhil, University of Waikato, Hamilton. Wiliam, D. (1992) Some technical issues in assessment: a user’s guide, British Journal of Curriculum and Assessment, 2(3): 11–20. Wiliam, D. (1994) Towards a Philosophy for Educational Assessment. Paper presented at the Annual Conference of the British Education Research Association, Bath, England. Wilkinson, L. (1993) The ‘Exscite’ experience: a school visit to an interactive science exhibition. Unpublished MSc thesis, University of Waikato, Hamilton. Williamson, V. (1954) Epic story of how large town supply farm developed, Town Milk, Journal of the New Zealand Milk Board, 4(4): 4–10, October. Wilson, K. (2002) The Treaty of Waitangi: preparing beginning teachers to meet the expectations of the new professional standards, Waikato Journal of Education, 8: 27–42. Winch, C. and Gingell, J. (1999) Key Concepts in the Philosophy of Education. London: Routledge.
234
References
Windross, S. (1995) Sound and fury: an attack on New Zealand science education is unscientifically hit and miss, NZ Listener, 60 (22 July). Wood, R. (1996) From my DNA to Darwin: teaching and learning in genetics. Unpublished PhD, University of Waikato Hamilton. Yager, R. and Tamir, P. (1993) STS approaches: reasons, intentions, accomplishments and outcomes, Science Education, 77(6): 637–58. Yeatman, A. (1994) The epistemological politics of postmodern feminist theorising, in A. Yeatman (ed.), Postmodern Revisionings of the Political. New York, NY: Routledge.
Index
235
Index
Titles of publications are in italics. accountability assessment 117–18 achievement, Maori students 177–9 active construction of knowledge 34 active learning 23–4 addressivity 57 affect, classroom interaction 106–7 affordance 48 ako Maori 157 alternative conceptions: energy 80; science 2, 17–21, 61–70, 99 Alvarez, A. 50 anecdoting as a pedagogical tool 102 animal, children’s conceptions of 18–19; teaching the concept 65–6 appropriateness of assessment 141 appropriation 45 artefacts for learning support 47–8 Asian students, cultural influences 45 Asoko, H. 40 assessment 116–43; and constructivist teaching 64; formative 121–33; Maori students 170–2, 178; quality 138–42; summative 133–8 assimilation policies and Maori 147–8 astronomy teaching 97–9 Augoustinos, M. 54, 56, 60 Australian Science Education Project (ASEP) 25 Ausubel, D. 21–2 Bakhtin, M. 57, 58 Barker, M. 20, 65, 67 beginning teachers development 110–12 Bell, B. 13, 20, 36, 41, 42, 65, 101, 109, 122, 133–4, 161, 169 Berger, P. 40
better learning 110, 134–5 bicultural science education 160–3; see also Maori culture and science teaching Biddulph, F. 5, 74 bilingual education 156, 160–3; see also te reo Maori biology, partially open investigative work 91–2 Bishop, R. 147, 148, 153–4, 158, 175–6 Black, P. 13, 122–3 Bruner, J. 21, 30–1 capacitance, context-based teaching sequence 83–5 Carey, S. 26 Carr, M. 35, 37, 63, 80, 81–2 Chang, W. 103 chemistry, context-based teaching 85–6; grade-related assessment 136 Cheng, M. 110 children see students children’s science see alternative conceptions Chinese culture, influence on learning 45 Chu, M. 45 Clarke, S. 131 classroom discussion 70, 99–107 classroom experiences, practical work 3, 90–2 classroom practice development, teachers 185, 188 Cobb, P. 41–2, 51 cognition, distributed 46–8; in discursive approach 54–5 cognitive development, teachers 187 collaborative relationships, teachers 187, 190 collective philosophy, Maori 158
236
Index
communicating scientists’ ideas 100–1 conceptual development 21–4 consequential validity of assessment 140 constructivism 24–39; and knowledge 35–9; teaching activities 63 consumer, as biological concept 65 content of science 78–82 context of learning 30–1 context-based teaching 82–9, 105 Cosgrove, M. 73 Cowie, B. 13, 122, 128, 132–3, 140, 141 Crooks, T. 139 cross-cultural science teaching 175 cultural artefacts, and social learning 44 cultural aspirations principle 157 cultural deficit theory and Maori underachievement 148–9 culturally preferred pedagogy 157 culture, influence on learning 45–6; see also Maori culture curriculum development 180–1, 191–201 curriculum implications of LISP projects 69–70 curriculum, science, and Maori 153–6, 159–63, 164–8, 200 Curriculum Studies module, teacher education 111–12 Dawson, C. 64 del Rio, P. 50 Devine, N. 174 dialogicality 57–8 disclosure in assessment 141–2 discursive learning 52–60 dispositional theory of thinking 46, 96–7 distributed cognition 46–8 Doppler effect, context-based teaching sequence 83–5 Driver, R. 23, 36, 38, 40, 91, 133 Duckworth, E. 64 Duit, R. 31 Durie, M. 179 Duschl, R. 139 Eames, C. 107 earth sciences, teaching sequences 66–7 earthquakes, teaching using Maori myth 172–3 Easley, J. 133 education accountability 117–18 education policy and Maori 147–9 education relationship metaphors 176
electromagnetism, context-based teaching 87 enculturation 45–6 energy, teaching and learning 7–10, 78–82 epistemological criticisms of constructivism 32–5 equity of assessment 140 ethnic stereotyping perception 150 Evening, L. 93 extended family principle 158 fairness of assessment 140 feedback to students 129 fidelity of assessment 141–2 Fijian students, inclusion 93 formative assessment 13–15, 118, 121–33 Francis, I. 95 Freyburg, P. 119 Fullan, M. 193 gender, and science teaching 93–5 Generative Learning Approach 28–9 Generative Model of Teaching 70–4 genetics, context-based teaching 87–8 gifted students, inclusion 95–6 Gilbert, Jane 32–3, 36–8, 46, 93–5 Gilbert, John 2, 12, 18, 20–1, 41–2, 134 Gillett, G. 53, 54, 56, 59, 60 Gipps, C. 142 girls, inclusion in senior science 93–5 Gitomer, D. H. 139 Glynn, T. 147, 148, 153–4, 158, 175–6 Gore, J. 113 grade-related criteria (GRC) assessment 135–6 Gribble, W. 86, 172 Gunstone, R. 64 Haami, B. 161–2 Haigh, M. 91 Hameed, H. 96 Happs, J. 66 Hargreaves, A. 193 Harlen, W. 122 Harré, R. 36–7, 53, 54, 56, 59, 60 Hattie, J. 123, 129 Hennessy, S. 45 Hesse, M. B. 98 Hewson, P. 22, 23 Hodson, D. 51 human action 48–9; as unit of analysis 52
Index inclusive pedagogies 92–6 innovations, and curriculum development 195 integration policy, Maori education 148 intelligibility of conception 22, 23 interactive formative assessment 124–9 Interactive Teaching Approach 7, 74–8 interviews as assessment 171–2 Jay, E. 46, 96 Jones, A. 83 kaupapa Maori 157–8 Kelly, G. 27 Kent, L. 170–1 kia piki ake i nga raruraru o te kainga 157–8 kiore rat, cultural significance 162–3 Kirkwood, V. 63, 79–82 kura kaupapa Maori 157 Lal, K. 136 language: and learning 52–60; Maori see te reo Maori; and scientific terms 99 Language, Culture and Science Education conference 170 Lave, J. 44–5 Leach, J. 40, 113 learning; as conceptual development 21–4; discursive 52–60; as distributed cognition 46–8; and enculturation 45–6; as a mediated action 48–9; as a situated activity 44–6; social 43–4 learning goals, assessment 118–21 Learning in Science Projects: Assessment 13–15, 120, 122; curriculum implications 69–70; Energy 7–10, 78–82; F1-4 1–4; and Maori 149–53; Primary 4–7, 119; Teacher Development 10–13, 41, 108–12, 134–5, 181–90 legitimate peripheral participation 45 Lemke, J. 55 LISP see Learning in Science Projects living world strand, science curriculum 166–7 Luckmann, T. 40 Mahuta 152 Maldives: contextual teaching 86–7; practical work 91; students’ thinking disposition 97
237
manageability of assessment 141 Maori: and curriculum development 164–8, 200; and education policy 147–9; and Pakeha 166–8, 175–6; philosophy 157–8; schools 156–7; and science education 149–54, 159–63, 164–79 Maori culture: and science teaching 160–1, 170–3; suppression of 147–8 Maori language see te reo Maori Maori students: assessment 170–2, 178; experiences 174–5; scientific attitude 149–51 Matthews, M. 32 May, S. 156 McKinley, E. 145, 149, 152, 154, 157, 161, 165–8, 169, 173, 174–5, 177–8, 179, 180 McPherson Waiti, P. 161, 169, 173 Messick, S. 140 metacognition 96–9 metaphors for educational relationships 176 Millar, R. 39, 63, 91 model-building approach, astronomy teaching 97–9 Mohamed, A. M. 86 Monk, G. 60 Moon, B. 113 Mortimer, E. 40 narrative as a pedagogical tool 102 national curriculum, teacher involvement 194 National Education Monitoring Project (NEMP) 178 negative affect, classroom interaction 106 novice teachers, development 110–12 Nuthall, G. 17, 45, 50, 59 O’Loughlin, M. 26–7, 55 5 Mataroa strand of science curriculum 166–7 ontological criticisms of constructivism 32–5 origin of universe, Maori culture 167 Osborne, R. 4–5, 28–9, 119 Oxenham, L. 100 Pacific Island students: inclusion 93; scientific attitudes 149–51 Pakeha, relationship with Maori 145, 166–8, 175–6
238
Index
partially open investigative work 91–2 Pea, R. 52 pedagogy 113–15 peer interaction and conceptual change 103 Perkins, D. 43–4, 46, 47, 48, 51, 52, 96 personal construct theory 27–30 personal constructivist learning 2–3, 28–30 personal development, teachers 182–4, 186–7, 188, 192–3 philosophy, collective, Maori 158 photosynthesis teaching package 67–8 physics, context-based teaching 86–7, 88–9 Piagetian constructivism 24–7 Pinar, W. F. 200–1 planned formative assessment 123, 125–9 plausibility of conception 22 Polynesian students, scientific attitudes 149–51 Porteous, D. 88 Posner, G. 22 post-colonial views of Maori education 174 power relationship, Pakeha and Maori 166–8, 175–6 practical work 3, 90–2 privileging of social language 59 professional development, teachers 182, 184–6, 187–8, 189, 191 Pryor, J. 127, 132 psychological basis, science curriculum 167 quality in educational assessment 138–42 Raghavan, M. 103 Raven, R. J. 25 Rawaikela, M. 93 reading texts as a pedagogy 101 relativism and constructivism 33 Roberts, M. 161, 162 Rodrigues, S. 85–6, 102 Rogoff, B. 51 Roth, W.-M. 35 Rowell, J. 63–4 Roychoudhury, A. 35 Saeed, S. 91 Salomon, G. 43–4, 47, 51
Schutz, A. 40 Science in the New Zealand Curriculum 97 science camps 173 science curriculum development 180–1, 191–201; and Maori 164–8, 200 science education and Maori 144–5, 149–54, 159–63, 164–79 science educators as curriculum shareholders 199–200 scientific attitudes, Polynesian students 149–51 scientific ideas, communication of 100–1 scientific terms see alternative conceptions Scott, P. 40, 58 self-assessment, formative assessment model 129 self determination principle 157 self-esteem, teachers 108–10, 185 sexual differences and science teaching 93–5 situated activity, learning 44–6 skills, assessment of 118–21 Smith, G. 153, 157, 176 social constructivism 40–3 social development, teachers 182, 184, 187, 188–9, 191–2 social interaction in classroom 105 social languages 58–9 social learning 43–4 sociocultural practices, workplace learning 107 sociocultural views of learning 43–52 socioeconomic mediation principle 157–8 Solomon, J. 31, 32 speech genres 58–9 Spolsky, B. 156 Stead, K. 149–52 Stewart, G. 178 students: alternative conceptions see alternative conceptions; as curriculum shareholders 199; scientific outlook 149–51; questions 6 Students with Special Abilities (SWSA) 95–6 summative assessment 118, 133–8 symbolic world, and learning 56 Taha Maori and the science curriculum 153–4 taonga tuku iho 157 Tasker, R. 20–1, 119
Index Taylor, I. 97–9 Taylor, P. 190 te reo Maori: decline 155, 158; and education 154–8, 159–63; science assessments 178; science curriculum 164–5, 168 Te TauAki Marautanga Pìtaiao He Tauira (science curriculum draft version) 165 teacher development 11–12, 108–12, 181–90; and curriculum development 191–7 teachers: as curriculum shareholders 197–8; primary 4–6; roles 68–9, 76 teaching activities, new, and teacher development 185 teaching strategies 6–7; see also constructivism; context-based teaching; practical work theorising of assessment 142–3 thinking apprenticeship 97 thinking, dispositional theory 96–7 thinking skills, assessment 119 Thorley, N. 22, 23 tino rangatiratanga 157 Tishman, S. 46, 96 Tittle, C. 128 Torrance, H. 127, 132 Torrie, I. 135–6 total immersion schools 156, 173
239
Treaty of Waitangi 146 trust: and assessment 141; teachers 188 underachievement, Maori 151–2, 173 universe origin, Maori culture 167 utterance 57–9 validity of assessment 139–40 voice 57 volcanoes, teaching using Maori myth 172–3 Vygotsky, L. 49 Waitangi, Treaty of 146 Waiti, P. 180 Walker, I. 54, 56, 60 Walkerdine, V. 167 wananga (science camps) 173 Wenger, E. 44–5 Wertsch, J. 49, 50, 52, 57, 58, 59 whanau (extended family) 158 White, R. 30 Wigglesworth, P. Z. 106 Wiliam, D. 13, 122–3, 141 Wittgenstein, L. 53–4 Wittrock, M. 28–9 Wood, R. 87–8 workplace learning 107 Yeatman, A. 33