Children’s definitional skills and their relations with metalinguistic awareness and school achievement

© 2004 Edizioni Carlo Amorewww.carloamore.it

Firera Publishing GroupLondon • Madrid • New York • Paris • Rome

European Journalof

School Psychology

A.Solombrino

Special issue

A.Solombrino

Special issue

European Journal of School Psychology

ISSN: 1723-7254



www.firera.com

Editor Board Sergio Salvatore – Department of Education, Psychological and Teaching Sciences, University of Lecce, Italy

Co-Editor Gioacchino Lavanco – Department of Psychology, University of Palermo, Italy

Editor AssistantMaria Beatrice Ligorio, University of Bari, Italy

Advisory Editor BoardPaola Bastianoni, University of Lecce, ItalyAlexandar Baukal, University of Belgrade, YugoslaviaMichele Cesaro, University of Salerno, ItalyHenny van der Meijden, University of Nijmegen, NetherlandMaria Francesca Freda, University of Naples “Federico II, ItalyAntonio Iannaccone, University of Salerno, Italy Marjaana Rahikainen, University of Helsinky, Finland

Board SecretaryChiara de Franchis, University of Lecce, Italy

Advisory BoardProf. Albert Bandura, Stanford University, CA, USAProf. Lucia Boncori, University of Rome “La Sapienza”, ItalyProf. Luigi Campagna, Polytechnic University of Milan, ItalyProf. Renzo Carli – University of Rome “La Sapienza”, ItalyProf. Micheal Cole – University of California, San Diego, USA Prof. Martin Covington –University of California, Berkley, USAProf. Lucia De Anna, IUSM, Rome, Italy Prof. Severino De Pieri, International School of Education, Venice, Italy


Prof. Franco Di Maria, University of Palermo, ItalyProf. Filip Dochy, University of Maastricht, NetherlandsProf. Bianca Gelli, University of Lecce, ItalyProf. Ana Gomes, University of Belo Horizonte, Brazil Prof. Barry Grant, University of Maryland, College Park MD, USAProf. Massimo Grasso, University of Rome “La Sapienza”, ItalyProf. Hubert J.M. Herman, University of Nijmegen, NetherlandsProf. Ivan Ivic, University of Belgrade, YugoslaviaProf. Willy Lens, Katholieke University of Leuven, BelgiumProf. Robert J. Marzano, Mid-continent Research for Education and Learn-ing (McREL) Institute, Colorado, USA Prof. Gianni Montesarchio, University of Rome “La Sapienza”, ItalyProf. Carlo Nanni, Salesian University of Rome, Italy Prof. Adele Nunziante Cesaro, University of Naples “Federico II”, ItalyProf. Nicola Paparella, University of Lecce, ItalyProf. Daniela Peja, RomeTre University, Rome, Italy Prof. Michele Pellerey, Salesian University of Rome, Italy Prof. Daniel Peraya, University of Ginevre, SwitzerlandProf. Luciano Pero, Polytechnic University of Milan, Italy Prof. Angela Perucca, University of Lecce, ItalyProf. Carlo Petracca, LUMSA University, Rome, ItalyProf. Andrew Pollard, University of Cambridge, United Kingdom Prof. Jean-Pierre Pourtois, University of Mons-Hainaut, BelgiumProf. Gaetano Quarta, University of Lecce, Italy †Prof. Marlene Scardamalia, University of Toronto, CanadaProf. Laura Sestito, University of Naples “Federico II”, Napoli Prof. Robert E. Slavin, John Hopkins University, Baltimore, Maryland, USA Prof. Robert J. Simons, University of Utrecht, NetherlandsProf. Gerry Stahl, Fraunhofer Institute for Applied Information Technology, FIT, GermanyProf. Jaan Valsiner, Clark University, Worcester, MA, USAProf. Pierre Vermersch, GREX, Paris, FranceProf. Gesualdo Zucco, University of Padova, Italy

GoalsThe journal plans to develop a holistic approach, addressing a comprehensive range of issues within the field of school psychology and the more general school context.

AudienceThe journal is intended for researchers (above all psychologists, but also edu-cators and sociologists) involved in scientific and applied research in educa-tional fields and practitioner psychologists working in school settings who are interested in strategies, intervention models, tools and data validated on a theoretical and empirical level according to the standards of international literature.

Focus of the Journal1. The journal aims to provide a vision of psychology as a methodology

for knowledge and intervention in and for educational contexts. In other words, research and psychological theory are a way of seeing, by means of which the phenomena of the school environment are investigated in order to provide resources to those involved in the educational process (above all teachers and managers). This underlying rationale differs significantly from a definition of psychology in terms of specific objects or phenome-na of study (e.g conceptual change, emotional behaviour disorder, teacher effectiveness, mathematics ability, school effectiveness, etc.).

2. The focus of attention is on intervention in a methodological rather than technical and operational sense. This means that importance is given to:

a) the analysis of school processes and problems;b) the relationship between, on the one hand, demand and, on the

other, the theoretical models, techniques and instruments of psy-chology;

c) the theoretical foundations, the heuristic validity and the education-al, organisational and social impact of school psychology.

3. Editorial policy intends to develop in two directions. Firstly, the journal

will provide an open platform for scientific debate, encouraging a spec-trum of models and points of view. In this regard EJSP will give guide-lines for basic methodological and formal standards and a shared code of practice, so that a dialectical approach can give differing stances their rightful value. Secondly, the journal intends to create for itself an original scientific direction, thus functioning as a cultural agent. In this direction the following points provide a starting point:

a) the contextual, contingent, organisational and social nature of the learning and teaching processes;

b) the consequent impossibility of separating the institutional and or-ganisational dimensions from the learning and teaching processes;

c) the importance of meta-cognition, deriving from the processes of emotional symbolisation mediating the relationship between indi-viduals, learning and contexts;

d) the need to ground professional practice and training in sound theo-retical and methodological foundations.

This rationale requires an interdisciplinary approach that on the one hand addresses the basic research agenda of psychology relating to different fields (e.g. social cognition, devel-opmental psychology, psychodynamic theory, instructional psychology, etc.) and on the other hand seeks interlocutors in the pedagogical sciences and more in general in the numerous fields of social and organisational thought (e.g. service management, organisational theory, anthropology, semiotics, philosophy, linguistics, sociology of welfare etc.)

Contributions 1. The journal welcomes different types of contributions: reports of empiri-

cal investigations; theoretical papers, literature reviews, case studies, book and article reviews, educational web site commentaries.

2. Submissions will be organised into four broad categories:

i) the school system: research and analysis concerning problems and processes of the school system, autonomy, assessment, social inclu-sion...);

ii) context and organisation: analysis of school organisation and policy;

relationship between educational proposals and user demand, rela-tionship between organisation and the learning and teaching proc-esses;

iii) learning models: psychological models relating to the understanding and management of the instructional and didactic processes;

iv) theory of technique and strategies of psychological intervention: design, imple-mentation and evaluation of psychological intervention; tools and methodology of professional action.

OrganisationThe three levels of organisation of the journal are:

1. International Advisory Board. The International Advisory Board has the func-tion of overseer. It develops the scientific directions of the journal and verifies the validity and implementation of quality standards. It is hoped gradually to increase the number of members of the board to form a solid group of distinguished scholars from the fields of international psychol-ogy and school psychology.

2. Editorial Board. The editorial board runs the journal implementing the guidelines defined by the International Advisory Board. It is responsible for developing the specific cultural direction of the journal. Members of the Editorial Board represent the various disciplinary areas of the journal’s editorial project.

3. Referees. Referees will be selected from among researchers and experts in the field addressed by the papers submitted.

Assessment of submissionsPapers submitted for publication will be examined first by the Editor along with a member of the advisory board. If the paper is considered suitable for publication, the editor will select two referees for a blind review. The journal will accept publication if both referees approve the paper. If there is disa-greement, a third referee will be called upon to review the paper. No papers will be published without the approval of at least two referees.

EUROPEAN JOURNAL OF SCHOOL PSYCHOLOGY Vol. 2, No. 1-2

CONTENTS

SPECIAL ISSUEWhat Students and Teachers Think about Learning: Contextual Aspects

Edited by Alessandro Antonietti

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15Alessandro Antonietti

Conceptions of teaching and learning at school and university: similarities, differences, relationships and contextual factors . . . . . . . 19Gillian M. Boulton-Lewis

On the nature of epistemologies: making explicit hidden assumptions through analysing instrument design . . . . . . . . . . . . . 39Theo J.H. Niessen, Jan D. Vermunt, Tineke A. Abma, Guy A.M. Widdershoven, Cees P.M. van der Vleuten

Teacher-child relationship: a meeting of minds . . . . . . . . . . . . . . .65Flavia Lecciso, Olga Liverta-Sempio, Antonella Marchetti

Effects of educational culture and teaching experience on teachers’ beliefs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83

Ottavia Albanese, Pierre-André Doudin, Caterina Fiorilli, Roberta Garbo

Approach to learning in the primary school: mathematics compared to language instruction . . . . . . . . . . . . . . .99Manuela Cantoia

Conceptions of learning and use of cultural media . . . . . . . . . . . . 127Sonia Perèz-Tello, Alessandro Antonietti, Olga Liverta-Sempio, Antonella Marchetti

Intelligence, educational practices and school reform: organisations change, representations persist. . . . . . . . . . . . . . . . . 149Felice Carugati, Patrizia Selleri

At the crossroads: the development of epistemological beliefs and historical thinking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169Liliana Maggioni, Patricia Alexander, Bruce VanSledright

ARTICLES

Creating new classroom culture via a virtual world: a case study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201Henny van der Meijden, Maria Beatrice Ligorio

School climate and well being in early adolescence: a comprehensive model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219Alessio Vieno, Massimo Santinello, Elena Galbiati, Massimo Mirandola

Children’s definitional skills and their relations with metalinguistic awareness and school achievement . . . . . . . . 239Gianluca Gini, Beatrice Benelli, Carmen Belacchi


Special issue

What Students and Teachers Think about Learning:

Contextual Aspects

edited by

Alessandro Antonietti

European Journal of School Psychology - Vol. 2, No. 1-2, 15-17 (2004)

15

INTRODUCTION

Alessandro Antonietti*

This special issue of the European Journal of School Psychology originated in the thematic session on Representations and naive conceptions of the learning-teaching process included in the Conference entitled Context, culture, intervention. What psychology for the future school? held in Lecce on June 20-22 , 2003. The papers presented by Italian researchers in that meeting have been integrated with further contributions by well-known scholars so to devise an international overview of the recent advances in the field.

The aim of the thematic session, as well as of the present issue, was to analyse how both students and teachers represent to themselves what occurs while they are, respectively, learning or teaching. This subject was faced by stressing the role of contextual factors. The two terms which define the spe-cificity of the topic – namely, representation and context – have to be clarified.

What does it mean that the main actors operating in schools and univer-sities can represent to themselves the process in which they are engaged? It means that they perceive – implicitly or explicitly – some psychological aspects of the setting in which they are involved and can reflect about them. For instance, they appreciate the intensity and the quality of the interpersonal relationships which link students and teachers and of the emotional reactions that they experience; develop opinions about the motivation which supports their efforts; share beliefs about the nature of the knowledge that they are asked to acquire and to elaborate; conceive personal ideas about the abilities that are required and about what facilitates or hinders the achievement of the goals. Notions such as attributions, naive conceptions, folk theories, personal epistemologies, social representations have been proposed to describe these subjective thoughts that learners and trainers construe in their mind. Leaving aside the differences existing among these notions, the common assumption

* Department of Psychology, Catholic University of the Sacred Heart, Milano, Italy.

Alessandro Antonietti

16

is that students’ and teachers’ behaviour is (at least partially) driven by what they think about the specific issues (affects, subject matters, cognitive strate-gies, skills, and so on) to which they are faced in school.

Are these kinds of representation only the products of the internal mental work of the individuals? The papers here presented try to show that vari-ous contextual factors intervene in modulating personal perceptions and be-liefs. After the general review initially provided by Gillian M. Boulton-Lewis and the analysis, carried out by Theo J.H. Niessen and his colleagues, of the implicit assumptions which lie under the methods and instruments usually employed in this area of research, Flavia Lecciso, Olga Liverta-Sempio and Antonella Marchetti highlight children’s and teachers’ reciprocal representa-tions of the attachment. Interpersonal relationships constitute a sort of ba-sic contextual framework sustaining education. Intersubjective relationships, however, are not symmetrical in school. Ottavia Albanese and her colleagues show as teachers’ expertise influence the representation of the intelligence and of its modifiability within school settings.

The subsequent papers move our attention toward an other aspect of the context: culture. Culture

is embedded in the disciplined taught at school: Manuela Cantoia com-pares children’s opinions concerning mathematics and linguistic education and finds that the subject matter orientates pupils’ perceptions, attributions and beliefs differently. Students interact with culture not only through the school disciplines, but also through several media that they use, or are ex-posed to, in everyday-life contexts. The paper by Sonia Pèrez-Tello and co-workers describes the links existing between conceptions of learning and the frequency of use of tools such as television, radio, computer, video- and audio-cassettes, as well as of books and magazines.

Institutions represents a third kind of context which is considered in this special issue. The contribution by Felice Carugati and Patrizia Selleri discusses the interplay which connects personal opinions, social tendencies, and politi-cal attempts to reform the school. The tensions – such as those considered by the above mentioned authors – which permeate people’s beliefs remind us that the representations in which we are interested are not static but, as shown by Liliana Maggioni, Patricia Alexander and Bruce VanSledright in the final paper, can change as a consequence of focused educational interven-

Introduction

17

tions.The set of articles which constitute this issue of the European Journal of

School Psychology give an account of different aspects of the representation of the learning-teaching process. They consider both students’ and teachers’ perspectives, both school and university, both general and domain-specific beliefs. We hope that they can provide readers a fruitful opportunity to draw an updated picture of this field of investigation.


19

CONCEPTIONS OF TEACHING AND LEARNING AT SCHOOL AND UNIVERSITY:

SIMILARITIES, DIFFERENCES, RELATIONSHIPS AND CONTEXTUAL

FACTORSGillian M. Boulton-Lewis *1

This paper is a discussion of the relationships between and within conceptions of teaching and learning in schools and universities. It becomes apparent that generally and statistically there is congruence between conceptions of learning and approaches to it and between conceptions of teaching and strategies at all levels and across contexts. There is also generally congruence between conceptions of teaching and learning. However when one looks closely at some studies, it is possible to determine pockets of dissonance in the relationships. This is important to recognise because it influences teaching and learning and outcomes for students and teachers. There are also similarities and differences across cultural contexts, discipline areas and levels of education.

THE STUDY OF CONCEPTIONS

Most of the conceptions of teaching and learning discussed in this paper are derived from phenomenographical research although a few studies have different theoretical perspectives or methodology. Some have operationalised the conceptions and developed inventories whilst others have combined interviews and questionnaires. Phenomenography is a qualitative methodology that is empirically based and aims to identify the different ways

* Faculty of Education, Queensland University of Technology, Australia.1 The author acknowledges the contribution to this work provided by Lynn Wilss. She has co-authored and undertaken research for some of the articles cited and provided valuable critical comment for this review.

Gillian M. Boulton-Lewis

20

(or variations in the ways) in which people experience, conceptualise, perceive or understand various phenomena (Marton, 1988). Marton and colleagues in Gothenburg have developed this approach over about 30 years. It was originally applied to understanding student learning and has since been used in other discipline areas.

The focus of the methodology is directed towards determining categories of conceptions of a phenomenon such as learning. The categories are derived from the data (usually individual interviews) and illustrated and supported by excerpts from them. The categories are descriptions of the ways in which a specific phenomenon appears to a group of people. They are usually presented in a hierarchy in which the descriptions are delimited structurally and referentially from each other. The presentation of the categories is often in a form described as an outcome space. A well known set of hierarchical categories of conceptions is that proposed by Marton, Dall’Alba & Beaty (1993) for learning by Open University students in the United Kingdom. They were as follows: A. Increasing one’s knowledge, B. Memorizing and reproducing, C. Applying, D. Understanding, E. Seeing something in a different way, and F. Changing as a person. In this outcome space all the interview data were reduced to six essential different ways in which the particular group of university students experienced learning. They were delimited and supported by quotes from the interviews.

Richardson (1999) highlighted some limitations of and warnings about phenomenography which include its reliance on participant and researcher interpretations, contextualisation of experiences, need for authentic communication, the possibility of different researchers arriving at different categories, and the fact that samples are usually small so that one cannot generalise from the results.

Nevertheless the methodology is a unique way of describing the range of different ways in which people experience a phenomenon and has been very useful, as explained below, in research in teaching and learning at different levels of education and in a range of contexts. As Sandberg (1994) suggested the more faithful we, as researchers, can be to individuals’ conceptions of an aspect of reality, the better we are able to understand learning, teaching and other kinds of human action within society.

This work originated in a Western context in universities but recently it has been undertaken in different cultural contexts and in schools. A decision has

Conceptions of teaching and learning at school and university

21

been made to sample only some of the research in conceptions in universities and focus where possible on work in schools.

STUDENTS’ CONCEPTIONS OF LEARNING

At universityThe basis for work in conceptions of learning began in Sweden with

university students (Saljo, 1975; Marton & Saljo, 1976a, 1976b). Students were asked to read parts of chapters in an educational textbook. The intention of the study was to find out what they understood of the text. After reading the students were questioned and their responses were analysed. The outcome of the analysis was four distinctively different ways of making sense of the text which were hierarchical and ranged from surface to deep approaches. Deep approaches led to understanding the author’s message and surface approaches were related to misunderstanding. This was the beginning of a series of studies and on the basis of the results Saljo (1982) claimed that there was a relationship between general experiences of learning (conceptions) and ways of reading a text (approaches), thus congruence. Similar relationships between conceptions and approaches to learning have been proposed by other researchers (e.g. Marton, 1988; Trigwell & Prosser, 1991; van Rossum & Schenk, 1984). In later work Saljo (1979) identified five qualitatively distinct ways of experiencing learning. These were expanded by Marton et al. (1993) to six conceptions as described earlier. The latter study with adult students in the United Kingdom produced an extra category – F. Changing as a person. The categories are posited as hierarchical and inclusive. The lowest category A is concerned with getting facts and information. B is different in that the focus is on getting information to reproduce it for a purpose. In C the focus is widened to include using and applying the information at some future date. These three lower categories, of conceptions of what constitutes learning, are all quantitative and surface in nature. The next three categories D, E and F are focused on the qualitative aspects of learning at a deep level and are concerned with understanding. D is understanding by comparing, contrasting and making meaning; E is seeing something in a new way as a result of learning, and F is a view that the effect of learning is actually to change the


22

person.In a longitudinal series of studies with Australian Aboriginal university

students across disciplines Boulton-Lewis and others (Boulton-Lewis, Marton, Lewis & Wilss, 2000a, 2000b, 2004; Boulton-Lewis, Lewis & Wilss 2001) found categories of conceptions of learning similar to, but a little different from, those proposed by Marton et al. (1993). With first year students in the study (2000a, 2000b) there were three main categories with subcategories. They were 1. acquiring knowledge (a. increasing and b. using), 2. understanding (a. to acquire, b. to use, c. to relate, analyse and elaborate) and 3. personal growth. For the most part, the strategies these students used did not match/were not congruent with their conceptions. Boulton-Lewis et al. (2001) analysed the data for two years of the study and identified a set of categories of core conceptions for each year. The conceptions for the second year for the group were similar to those for the first year but two new conceptions were identified. These were 4. to see something differently and 5. a change in thinking and understanding. On an individual basis there were changes in some students’ awareness of learning. In analysing the data for the three years (Boulton-Lewis et al., 2004) it was decided after further consideration that the results could be described in a similar way to those identified by Marton et al. (1993) although there were still minor differences, including an emphasis on understanding, even with the lowest conception. For some students, contrary to findings in Sweden and the United Kingdom, there was dissonance between conceptions and approaches to learning. Most of this dissonance was caused by students voicing higher level conceptions than the strategies they used for learning. Over the 3 years of the study one third of the students exhibited consistently dissonant conceptions and strategies. The use of strategies that were at a lower level than their conceptions constrained their learning to a certain extent although this was compensated for by their strong motivation to obtain a qualification. The question of dissonance is addressed further in Boulton-Lewis et al. (2003). This group of students generally held lower level quantitative conceptions of learning and related strategies than other university students; however there were some moves to higher level conceptions and strategies over three years of study.

Similar studies have been undertaken with groups of university students from other cultural backgrounds. For example with distance learners at the University of the South Pacific, Mugler and Landbeck (2000) found what


23

could be interpreted as `levels of understanding’. Most students explained understanding as `knowing the content’ or `knowing how to do something’ while some stated that understanding was `knowing the meaning of subject matter’. This is similar to the three kinds of understanding described by Boulton- Lewis et al. Dahlin and Regmi (1995) interviewed both Nepali university students and secondary school students. They found conceptions similar to the original six described by Marton et al. (1993) but with different levels of memorising, more emphasis on meaning, more linkage between understanding and memorising and conceptions of the results of learning as `a change of behaviour, a change of consciousness or understanding, or change in both’. These last `change’ conceptions are a little like the 4th and 5th conceptions found for Aboriginal students (2001). Dahlin and Regmi proposed that culture does not determine the content of the learning experience in an absolute sense but that it does influence those aspects of the experience that are accentuated. Meyer and Boulton-Lewis (1999) operationalised conceptions of learning and other associated aspects in an inventory (RoLI) which was used with Australian, Indonesian and South African students. The results indicated that conceptions of learning are associated with cultural and experiential factors.

Eklund-Myrskog (1998) found similarities and differences between students’ conceptions of learning in two different educational contexts; student nurses and car mechanics. Differences between programs were greater than within programs with only two conceptions common to both student groups. The majority of the nursing students had a qualitative view of learning whilst the majority of the car mechanic students had a quantitative view. All students however had more developed conceptions at the end of their course. She states that the results showed that to some extent conceptions were contextually dependent.

From a different perspective Tynjala (1997) investigated the conceptions of the dimension of the learning process held by educational psychology students in a Finnish university. The set of seven conceptions that she proposed are interesting in that they are more concerned with how the student learns and what the student does to learn than what the student believes learning to be. These conceptions range from considering learning as an externally determined process to a creative process, are semi-hierarchical and not necessarily mutually exclusive at the individual level. Of course even with


24

a hierarchical set of conceptions, where students are then allocated to the category that fits them best, it is still likely that an individual will hold a range of conceptions.

The studies described above provide conceptions of learning across countries, courses and dimensions. Despite differences in emphasis and numbers of conceptions of learning in different contexts there is overlap and remarkable consistency with the original conceptions proposed by Marton et al. (1993).

They all range from quantitative conceptions of learning usually associated with surface strategies to qualitative conceptions which are associated with deep strategies and an intention to understand. Marton et al. (1997) suggested that the structure of the experience of learning in formal settings may be more or less universal. However this may simply reflect the culture of such institutions. Entwistle (1997) believes that as students progress through a course their conceptions will generally move from one of acquiring `discrete packages of information’ to one that constitutes change in themselves and the world around them and this seems to be borne out by most of the studies described above.

At schoolIn a written answer study with 150 14-16 year old Nigerian secondary

students, Watkins and Akande (1994), found four clear categories of descriptions of learning which were learning as: increasing knowledge (held by 47.4% of the respondents), memorising and reproducing (8.9%), applying (17%), and understanding (26.7%).

They suggest that these four categories were congruent with the first four described by Marton et al. (1993). In summary they proposed that the `quantitative memorising conception of learning and the conception of learning as understanding are identifiable in the responses of these Nigerian secondary school students’.

They also cite other Nigerian studies (Omokhodion, 1989, Ehindero, 1990) which found similar distinctions between reproductive and meaning oriented conceptions of learning. Watkins (1996) undertook further research with 20 junior and 24 senior secondary school students in Hong Kong with an emphasis on developmental aspects of learning. He described four stages


25

of intentions to learn which he linked with strategy, assessment, locus [of control], metacognitive [developmental] level and workload.

The stages of intentions, and associated strategies in parenthesis, were 1. achievement through reproduction (rote memorising);2. achievement through reproduction (rote learn important things);3. achievement through reproduction (understand then memorize);4. understanding and achievement (combine understanding and memo-

rising OR focus on understanding). He made some important points about this sequence when he says the

Chinese skill at memorising and the focus on it continue until the early years in secondary school until the amount to be learned becomes excessive so that at the stage 2 level students only memorised important things. By stage 3 they were combining memorising and understanding. He proposed that other factors influencing this progression were problems with the English language, levels of cognitive development, assessment rewards and self confidence.

A detailed analysis of continuities and discontinuities in learning by Chinese school children aged 12 to 18 years was undertaken by Marton, Watkins and Tang (1997). They found that the data could be depicted in an outcome space which showed four distinct ways of experiencing learning. These ranged from two conceptions of committing to memory (words or meaning) to two conceptions of learning as understanding (the meaning or a phenomenon).

There is a clear distinction between the two lower conceptions which are focussed on committing information to memory and the two higher conceptions which are concerned with meaning. The object of learning also ranged from words to meanings, to phenomena. The outcome space included a temporal facet to each of these ways of experiencing which moved from acquiring, to knowing, to using. For example in memorising words the first step was to memorise, the next step was to remember and the final stage was to reproduce. In understanding meaning the sequence was to first gain understanding, then remember meaning and finally to be able to do something, do something differently or do something different. There are some minor variations in these results from other categories of conceptions in that understanding precedes memorising in the temporal facet of understanding meaning.

Dart, Burnett, Purdie and Boulton-Lewis (2000) found a relationship


26

between qualitative conceptions and deep learning and quantitative conceptions and surface learning in a group of high school students in Australia. However, they also found a positive relationship between quantitative conceptions of learning and a deep approach and that the conceptions and approaches were affected by context.

Klatter, Lodewijks and Arnoutse (2001) described learning conceptions, or beliefs, about different aspects of learning, held by students in the final year of primary school in the Netherlands. They firstly interviewed 27 students about their learning and then a questionnaire was developed to determine the interrelatedness of beliefs about selected aspects. They implemented the questionnaire with 367 sixth graders. In the qualitative part of the study they found different beliefs about five aspects of learning; purpose of school, learning orientation, regulation, functional learning, and developmental learning.

These conceptions are different from the usual phenomenographic ones and the authors argue that they have redefined a learning conception as a multidimensional construct to include cognitive and affective components on one dimension. In the qualitative part of the study they asserted that three different types of learning conceptions could be identified ‘…a restricted learning conception characterised by a relatively low commitment to different kinds of beliefs, a functional learning conception, characterised by ego-orientation and external regulation for future success, and a developmental learning conception which is characterised by emphasis on personal growth and a deep approach to the learning process.’

They also found as one of their results that young pupils are not used to talking about learning as such but became more enthusiastic when they realised their discussions were confidential and that they could say whatever they liked. This can also be the case with university students who have not thought about learning explicitly. Boulton- Lewis et al. in their studies found that some of the Aboriginal students were not able to talk at length about learning until the second or third year of the study and spoke more easily to an Aboriginal research assistant who was not directly connected to their lecturers.

Purdie and Hattie ( 2002) developed an instrument from qualitative data, obtained with Australian and Japanese students, to assess conceptions of


27

learning (COLI) which they implemented with Australian (Indigenous and non- Indigenous), Malaysian and American ( Caucasian and African) high school students.

They found a relationship between self reported higher academic achievement and endorsement of all six conceptions of learning and not much support for two qualitatively different sets of conceptions, that is deep and surface, particularly with Asian students.

They also raised the issue of the importance of motivation in learning and in determining what is important, despite conceptions and the strategies students possess or choose to use. They proposed six conceptions as follows:

a) gaining information;b) remembering, using and understanding information;c) duty;d) personal change;e) a process not bound by time and place; f) social competence. Conceptions a, b and d are similar to those identified in other studies

and constitute experiences of what learning [cognitively] is thought to be. They fit with the phenomenographic view of conceptions where the concern is with experiences of a phenomenon in one dimension, in this case the cognitive aspect. Conceptions c, e and f are concerned with different aspects of learning; c) with affect, d) with temporal factors and f) with social factors. Hence, whilst they are all justifiable ways of thinking about learning they do not fit together well as one overall hierarchical set of conceptions.

In the research in conceptions of learning in school and university one can identify a developmental perspective. The studies with school students mostly show two distinct sets of conceptions concerned with either memorising or understanding, often with the emphasis on memorising of one kind or another. The conceptions for university students are augmented by more advanced qualitative conceptions which encompass seeing something in a different way or actually changing as a person. These more advanced conceptions are not always apparent at the beginning of university study for non-Western students or for those such as the Aboriginal students who, in many cases have had a less than satisfactory secondary school experience. It is suggested that the apparent developmental sequence for conceptions of


28

learning and congruent strategies is affected by context, general cognitive development, by acquisition of content knowledge and by challenges to learn more about learning. Generally students who hold higher conceptions of learning and approaches can orchestrate the whole range depending on the context and their motivation. Students who hold only lower level conceptions and approaches are limited to using those.

LECTURERS’ CONCEPTIONS OF STUDENT LEARNING

Bruce and Gerber (1995) conducted a study of lecturers’ conceptions of student learning in university. Unfortunately they did not link it empirically to their conceptions of teaching although they hypothesised about the relationships. They found six conceptions of learning (abbreviated) as follows:

1. acquiring knowledge; 2. absorption of knowledge to explain or apply; 3. developing thinking skills and the ability to reason; 4. developing the competencies of beginning professionals; 5. changing attitudes, beliefs and behaviours;6. a participative pedagogic experience. They suggest that the first and last categories may be due to personal

and contextual factors. The first because some of the participants were new to university teaching and the last because there was an emphasis on staff development for learning and teaching in the university where the study was conducted. They assert that the remaining categories are similar to those for teaching and learning proposed by others. They suggest that it is likely that lecturers’ conceptions of teaching and learning are related. This is generally likely to be true but there will also probably be dissonance as found in the school study by Boulton-Lewis et al. (2001) described later.

Newton, Newton and Oberski (1998) investigated learning and conceptions of understanding in history and science held by lecturers and graduates. They argue that understanding is a significant indicator of the quality of student learning. It can allow learners to explain, justify and predict; recall or reconstruct information; enable independent interaction with the world and facilitate the construction of further understanding. It is a component of the


29

higher levels of conceptions of learning proposed by Marton et al. (1993). Newton et al. (1998) found two conceptions of understanding in science; a capability to apply laws, and acquisition of a mental structure to explain these. The main conception in history was the reconstruction of events. Some of the lecturers’ and students’ conceptions were similar. However they make the point that when the students’ conceptions do not match those of the lecturers there is potential for learning difficulties because, for example, the lecturer may be aware of knowledge that the students barely understand or, on the other hand, it may be easy for the students to meet the lecturers’ expectations without understanding.

They suggest that students and lecturers should make explicit their awareness of understanding in order for some kind of shared meaning about learning to be developed.

CONCEPTIONS OF TEACHING

At universityKember (1998) summarised 14 studies of conceptions of teaching in higher

education. These included, among others, work by Martin and Balla (1991), Dall’Alba (1991), Samuelowicz and Bain (1992), Gow and Kember (1993), and Trigwell, Prosser and Taylor (1994). They were mostly phenomenographic studies. Kember was able to classify the studies into five categories which he then reduced to three based on teacher (lecturer) orientation. These were ‘teacher centred/content oriented’ and ‘student centred/learning oriented’ with an intermediate category that he described as ‘student teacher interaction/apprenticeship’. It is usually assumed that the student centred/learning oriented conception is the most effective for learning however it depends in part how this is operationalised. If students can be supported so that they take control of their learning and hence construct their own knowledge then such an approach should lead to deep learning. However, if students lack the prerequisite skills and knowledge, or motivation to behave in this way, then they can be ineffective as learners and become very frustrated. In such cases they probably need more scaffolding of their learning or, initially, a teacher centred/content oriented approach.


30

It is assumed that teachers’ conceptions of teaching will be aligned with, thus congruent with, their teaching approaches and strategies. For example Trigwell and Prosser (1996) confirmed statistically the congruence between intention and strategy in a study with first year science teachers. They suggested however that content and approach may be context dependent and that academic development focussing only on strategies is unlikely to be successful unless there is also a focus on the intentions associated with the strategy.

At schoolMost of the research in conceptions of teaching has been undertaken

in universities however two studies in secondary schools are summarised here, one in China and one in Australia. The second study also investigated the relationship between the teachers’ conceptions of teaching and their conceptions of student learning.

Lingbao and Watkins (2001) identified and assessed the conceptions of teaching held by secondary school physics teachers in China through interviews with 18 teachers, aided by episodes of classroom teaching processes, and an inventory developed from the interviews and used with 450 teachers. They identified five lower order conceptions of teaching which included knowledge delivery, exam preparation, ability development, attitude promotion and conduct guidance. These could be grouped in a higher order framework which focussed on moulding at the lower level to cultivating at the higher level. The results focus on more that one dimension and are therefore different from but similar to the usual conceptions of teaching described in the Western literature. It is proposed that this is probably due to cultural factors which emphasise an affective, moral view of teaching and the context of the school with an emphasis on examinations. They found that neither teaching experience nor gender were related to the conceptions. Most of the teachers responded positively to all the conceptions of teaching which implied that they held multiple conceptions. They explained why they shifted their conceptions during interviews by the fact that they had ideal conceptions about teaching and practical conceptions reflecting the school environment. The authors argued that this is evidence that teachers might support conflicting conceptions. Boulton-Lewis, Smith, McCrindle, Burnett and Campbell (2001)


31

investigated 16 secondary school teachers’ conceptions of teaching and student learning. They found four categories of the way in which teachers thought about teaching and student learning, the related strategies, and a focus on either content or student. The conceptions of teaching were: transmission of content/skills, development of skills/understanding, facilitation of understanding, and transformation (of students). The focus of these ranged from teaching content to change in students. Their conceptions of student learning were: acquisition and reproduction of content/skills, development and application of skills/understanding, development of understanding, and transformation of learners. Their conceptions of student learning also ranged from a focus on learners to a focus on content. Each teacher was then allocated to the category most typical of their conceptions of teaching and learning respectively. There was a lack of congruence in some cases between the conceptions they held. Notably there were four teachers (1/4 of the sample) whose conceptions of teaching and learning were not congruent. These four teachers held more sophisticated conceptions of teaching than their conceptions of learning. For example one teacher believed that teaching meant development of skills and understanding but that learning was acquir-ing and reproducing content and skills. Another teacher held a similar con-ception of learning but was concerned with student-teacher interaction to facilitate student understanding. The other two teachers believed that teach-ing was a process of facilitating student understanding through interaction but believed that learning was a matter mostly of development and applica-tion of the skills and understanding. These four teachers had higher goals for their teaching than for the outcomes of student learning. The inconsistencies displayed by these teachers suggest that one cannot assume that teachers’ conceptions of teaching will necessarily be aligned with their conceptions of student learning.

RELATIONSHIPS BETWEEN CONCEPTIONS ANDAPPROACHES TO LEARNING AND TEACHING

Cano (2004, submitted) has undertaken a detailed study of consonance and dissonance in European secondary school students’ experience of learning. A sample of 1012 students from grades 7 to 12 were asked to respond to open


32

ended questions and undertake the LPQ (Learning Process Questionnaire: Biggs, 1987) which measures surface motive and strategy, deep motive and strategy, and achieving motive and strategy. He found consonant and dissonant patterns in approaches to learning as well as in learning conceptions and strategies. There were two kinds of consonance (basic and complex) and two kinds of dissonance (negative and positive). The patterns of responses were significantly related to differences in performance. Positive dissonance (reproducing conceptions and higher level organisation/elaboration strategies) and complex consonance (deep conceptions and strategies) were associated with better results.

Studies by Boulton-Lewis and others, described above, found dissonance in the learning of some Aboriginal university students’ conceptions and approaches to learning and in teachers’ conceptions of teaching and learning. The dissonance for Aboriginal students was of the kind that Cano (2004) described as negative, that is students held higher level conceptions than strategies, however they managed to succeed through enormous effort. More congruence between conceptions and strategies consonance would have made their studies more rewarding.

In a complex study of the relationship between university assistants’ conceptions of knowledge, learning and instruction Buelens, Clement and Clarebout (2002) concluded that a relative conception of knowledge goes with a meaning-oriented conception of learning as well as with a student-oriented conception of instruction. On the other hand an absolute conception of knowledge is statistically related to reproduction–oriented learning and teacher-centred instruction. They also proposed that student-oriented conceptions of instruction are present early in a teaching career rather than necessarily being the product of experience and development popular in the literature.

Using interviews and a questionnaire with a sample of business lecturers, Murray and MacDonald (1997) found, that their expressed attitudes and beliefs about teaching were not translated into teaching strategies and beliefs. Most of the lecturers’ considered themselves to be facilitators of student learning or student supporters. However this was inconsistent with their views of lectures and tutorials which were to disseminate information, and apply or check knowledge or understanding. They refer to this disjunction, that is of beliefs not translated into teaching strategies and methods, in terms used by


33

Samuelowicz and Bain (1992) as one of the mysteries of higher education. They also liken it to work by Argyris and Schon (1978) who described the gap between espoused theory and theory in practice. It appeared also that the lecturers held multiple conceptions of their roles. They suggested that the differences between the ideal and the practical might have been caused by contextual factors such as large numbers of students or shortage of resources.

Trigwell and Prosser (1996) found strong relations between conceptions of teaching and approaches to teaching for a sample of physics and chemistry lecturers in Australia although relations between conceptions of teaching and learning were not so strong. It is apparent from the studies described above that there is usually a strong relationship between conceptions of learning and teaching and the approaches that lecturers and students adopt. However in some cases there is dissonance and in these situations the teaching or learning is usually not as effective.

SIMILARITIES, DIFFERENCES, AND CONTEXTUAL FACTORS

It might appear from this review that the results are inconsistent and confusing. There is certainly a complicated mixture of methodologies and results. However, despite that, common themes emerge across levels of education and cultures. In addition, in most of the studies of conceptions of learning and teaching, a range of related or overlapping categories of description have been identified.

In the studies of conceptions of learning at university there are two major perspectives, that is a quantitative view and a qualitative one. There are differences in some of the studies however between the conceptions held by students from different disciplines or cultural backgrounds. In the work dealing with processes of learning there is a demarcation, between considering learning as externally determined or as a creative process, which fits with the quantitative and qualitative conceptions of what learning is.

In the limited research in conceptions of learning held by school students there is congruence with the conceptions of learning found for university students with distinctions between a focus on reproduction or meaning


34

making but usually without the higher level qualitative conceptions.Lecturers’ conceptions of student learning fit to a certain extent with

student conceptions however there are some discipline differences.There are a large number of studies of conceptions of teaching at

university some of which have been reduced by Kember (1998) to 3 major groupings which focus on teacher and content, or student and learning, with an intermediate category of student teacher interaction.

The research on conceptions of teaching in schools produces lower and higher level conceptions with congruence generally between them and learning. However, here there are also differences due to cultural contexts and between beliefs and action.

IMPLICATIONS FOR TEACHERS AND STUDENTS

Because the research described above suggests that students will hold a range of conceptions of learning that will influence their performance there is a need for teachers to be aware of these, of their likely developmental sequence, and how they might be affected by culture and context.

Teachers also need to be explicitly aware of their own conceptions of teaching and student learning and try to ensure that there is congruence between them, or alternatively be flexible depending on the context and student requirements. It would be helpful if teachers made their perspectives on teaching and learning as explicit as possible to students and explained their reasons for them.

They will also be more effective if they plan their teaching so that conceptions of teaching and learning, and strategies and expectations are congruent or aligned as described by Biggs (1996).

Students should be helped to become aware of their conceptions of learning and the relationship between these and approaches to learning. They should ideally develop a range of deep and surface strategies which they can control in a metacognitive fashion to regulate their own learning depending on the context and their motivation and goals.

Basically it is suggested that we must help students at all levels to learn about learning. It would also be useful if they could be helped to understand the kind of teaching that they are experiencing to decide whether this fits


35

with their needs or whether they would prefer to learn in some other way, hence taking control of their own learning and the learning environment as far as possible.

CONCLUSION

The research in conceptions makes it clear that, whilst there may be accepted better ways of learning and teaching, students and teachers do not always accept or operate with these.

The suggestions above assume a strong commitment on the part of students and teachers to embracing better ways of teaching and learning however it is important to realise that there will be a range of other beliefs and expectations. Such a realisation is part of the way to being able to deal effectively with such beliefs, expectations and behaviour in order to improve learning for most students.

REFERENCES

Argyris, C & Schon, D. (1978). Organisational learning: A theory-of-action perspective. Reading, Massachusetts: Addison-Wesley.

Biggs, J.B. (1987). Learning process questionnaire. Melbourne: Australian Council for Educational Research.

Biggs, J.B. (1996). Enhancing teaching through constructive alignment. Higher Education, 32, 1-18.

Boulton-Lewis, G.M., Marton, F., Lewis, D. & Wilss, L. (2000a). Learning in formal and informal contexts: Conceptions and strategies of Aboriginal and Torres Strait Islander university students. Learning and Instruction, 10, 393- 414.

Boulton-Lewis, G.M., Marton, F., Lewis, D., & Wilss, L.A. (2000b). Aboriginal and Torres Strait Islander university students’ conceptions of formal learning and experiences of informal learning. Higher Education, 39, 469- 488.

Boulton-Lewis, G. M., Marton, F., Lewis, D.C. & Wilss, L.A., (2004), A


36

longitudinal study of learning for a group of indigenous Australian university students: Dissonant conceptions and strategies. Higher Education, 47(1), 91-112.

Boulton-Lewis, G.M., Lewis, D. & Wilss, L. (2001). Changes in conceptions of learning for Indigenous Australian university students. British Journal of Educational Psychology, 71, 327-341.

Boulton-Lewis, G.M., Wilss, L.A., & Lewis, D. (2003). Core conceptions of learning and ways of going about learning: Apparent dissonance for Indigenous Australian university students. Studies in Higher Education. 28 (1), 79-89.

Boulton-Lewis, G.M., Smith, D., McCrindle, A.R., Burnett, P.C. & Campbell, J. (2001). Secondary teachers’ conceptions of teaching and learning. Learning and Instruction, 11(1), 35-51.

Buelens, H., Clements, M. & Clarebout, (2002) University assistants’ conceptions of knowledge, learning and instruction, Research in Education, 67, 44- 57.

Bruce, C. & Gerber, R., (1995). Towards university lecturers’ conceptions of student learning. Higher Education, 29, 443-448.

Cano, F. (2004, submitted), Consonance and dissonance in students’ learning experience. Learning and Instruction.

Dahlin, B. & Regmi, M.P. (1995). Conceptions of learning among Nepali students. Paper presented at the 6th European Conference on Learning and Instruction in Njimegen, the Netherlands, August 26-31.

Dart, B.C., Burnett, P.C., Purdie, N. & Boulton-Lewis, G. (2000). Students’ conceptions of learning, the classroom environment, and approaches to learning. The Journal of Educational Research, 93(4), 462-474.

Eklund-Myrskog, G. (1998). Students, conceptions of learning in different educational contexts. Higher Education, 35, 299-316.s

Entwistle, N. (1997). Contrasting perspectives on learning. In F. Marton, D. Hounsell, & N. Entwistle (Eds.), The experience of learning: Implications for teaching and studying in higher education. Edinburgh: Scottish Academic Press.

Kember, D. (1998). Teaching beliefs and their impact on students’ approaches to learning. In B.Dart & G. Boulton-Lewis, Teaching and learning in higher education. Melbourne, Australia: Australian Council for Educational Research.

Klatter, E.B., Lodewijks, H.G.L.C., & Arnoutse, C.A.J. (2001). Learning


37

conceptions of young children in the final year of primary education. Learning and Instruction, 11(6), 485-516.

Lingbao, G. & Watkins, D. (2001). Identifying and assessing the conceptions of teaching secondary school physics teachers in China. British journal of educational psychology, 71, 443-469

Meyer, J.H.F. & Boulton-Lewis, G.M. (1999). On the operationalisation of conceptions of learning in higher education and their association with students’ knowledge and experience of their learning. Higher Education Research and Development, 18(3), 289-302.

Marton, F. (1988). Describing and improving learning. In R. Schmeck (Ed.) Learning strategies and learning styles. New York: Plenum.

Marton, F. (1988). Phenomenography: Exploring different conceptions of reality. New York: Praeger.

Marton, F. & Säljö, R (1976a). On qualitative differences in learning I – Outcome and process. British Journal of Educational Psychology, 46, 4-11.

Marton, F. & Säljö, R (1976b). On qualitative differences in learning II – Outcome as a function of the learner’s conception of the task. British Journal of Educational Psychology, 46, 115 -127.

Marton, F., Dall’Alba, G., & Beaty, E. (1993). Conceptions of learning. International Journal of Educational Research, 19, 277-300.

Marton, F. & Booth, S. (1997). Learning and awareness. Erlbaum, New Jersey.Marton, F., Watkins, D. & Tang, C. (1997). Discontinuities and continuities in

the experience of learning: An interview study of high-school students in Hong Kong. Learning and Instruction, 7 (1), 21-46.

Mugler, F. & Landbeck, R. (1997). Learning, memorisation and understanding among distance learners in the South Pacific. Learning and Instruction, 10, 179- 202.

Murray, & MacDonald, (1997).Newton, D.P., Newton, L.D. & Oberski, I (1998). Learning and conceptions

of understanding in History and Science: Lecturers and new graduates compared. Studies in Higher Education, 28 (1), 43-58.

Purdie, N. & Hattie, J. (2002). Assessing students’ conceptions of learning. Australian Journal of Educational and Developmental Psychology, 12, 17-32.

Richardson, J.T.E. (1999). The concepts and methods of phenomenographic research. Review of Educational Research, 69(1), 53-82.

Säljö, R. (1975). Qualitative differences in learning as a result of the learner’s conception


38

of the task. Acta Universitatis Gothoburgensis, Göteborg.Säljö, R. (1979). Learning in the learner’s perspective. I. Some common-sense

conceptions. Reports from the Department of Education, Göteborg University, No.76.

Säljö, R. (1982). Learning and understanding: A study of differences in constructing meaning from a text. Acta Universitatis Gothoburgensis, Göteborg.

Samuelowicz, K & Bain, J.D. (1992). Conceptions of teaching held by academic teachers. Higher Education, 34, 93-111.

Sandberg, J. (1994). Human competence at work: An interpretative approach. BAS, Göteborg.

Trigwell, K. & Prosser, M. (1991). Relating approaches to study and quality of learning outcomes at the course level. British Journal of Educational Psychology, 61, 265-275.

Trigwell, K, Prosser, M. & Taylor, P. (1994). Qualitative differences in approaches to teaching first year university science. Higher Education, 27, 75-84

Trigwell, K. & Prosser, M. (1996). Congruence between intention and strategy in university science teachers’ approaches to teaching, Higher Education, 32 (1), 77-87.

Tynjälä, P. (1997). Developing education students’ conceptions of the learning process in different learning environments. Learning and Instruction, 7(3), 277-292.

van Rossum, E.J. & Schenk, S.M. (1984). The relationship between learning conception, study strategy and learning outcome. British Journal of Educational Psychology, 54, 73-83.

Watkins, D. (1996). Hong Kong Secondary School Learners: A developmental perspective. In D.A. Watkins & J.B.Biggs (Eds). The Chinese learner: Cultural, psychological and contextual influences. ACER: Melbourne, Australia.

Watkins, D. & Akande, A. (1994). Approaches to learning of Nigerian secondary school children: Emic and etic perspectives. International Journal of Psychology, 29 (2), 165-182.


39

ON THE NATURE OF EPISTEMOLOGIES:MAKING EXPLICIT HIDDEN ASSUMPTIONS

THROUGH ANALYSING INSTRUMENT DESIGN

Theo J.H. Niessen*, Jan D. Vermunt**, Tineke A. Abma***, Guy A.M. Widdershoven***, Cees P.M. van der Vleuten*1

Many researchers from different disciplines have taken interest into the domain of personal epistemology. Although flourishing, the field is dispersed with many dis-parate labels which are scarcely clarified. This is problematic since it hampers the development of the domain and it ignores the fact that researchers of different ed-ucational communities impose different meanings on, and inflict different default assumptions to, the nature of the construct. To address these assumptions, in this article a six-dimensional framework is developed which focuses on the design of the instrument. It is contested that the instrument design reveals cues about the assump-tions behind the nature of the construct. These cues can be placed along these di-mensions. Next 30 instruments with different designs were gathered. Some of these instruments are scaled on the developed dimensions. It is concluded that the discus-sion on the nature of the constructs is well facilitated and drawn to a more explicit level by focusing on the instrument design.

“There are in fact a number of reasons for selecting a methodological approach, but one’s decision often expresses values about what the world is like,

how one ought to understand it, and what the most important threats to that understanding are.”

(Firestone, 1987)

* Maastricht University, Department of Educational Development and Research, The Netherlands.** Leiden University, ICLON – Graduate School of Education, The Netherlands.*** Maastricht University, Department of Healthcare Ethics & Philosophy, The Netherlands.1 Correspondence address: Theo Niessen, Maastricht University, Department of Educational, Development & Research, PO Box 616, 6200 MD, Maastricht, The Netherlands, Telephone: 0031-43-3881055, Fax: 0031-43-3884140, E-mail: [email protected]

Theo J.H. Niessen et al.

40

INTRODUCTION

Research on epistemologies is flourishing. Many researchers from differ-ent disciplines have taken an interest into the domain of personal episte-mology. The interest educational psychology has taken into epistemological issues is of reasonably recent date and started in the sixties of the previous century with the seminal work of Perry (Perry, 1968, 1988). Epistemology in general is concerned with questions like: What is the nature of knowledge?, and: How does a person come to know? Students and teachers show develop-ments in their ideas about these issues which in turn tend to influence their behaviour. Research on epistemology has thus made important contributions to education by identifying epistemology as a category of informal knowl-edge that may play a role in students’ knowledge, reasoning, study strategies, and participation (Hammer & Elby, 2002).

Though promising, this accomplishment partly masks the fact that the field lacks a unitary conceptual framework. The domain is dispersed with many disparate labels to indicate someone’s epistemological commitment: epistemological beliefs, epistemological thinking, epistemic cognition, episte-mological theories, ways of knowing, epistemological posture, epistemologi-cal reflection, epistemological stance, epistemological resources and episte-mological repertoires (Hofer & Burr, 2001; Hofer & Pintrich, 1997; Hofer & Pintrich, 2002; Pintrich, 2002).

This woodland of different terms hampers the further clarification and advancement of the epistemological domain (Hammer & Elby, 2002). More-over, researchers within the epistemological domain ground themselves in research that derives from different world hypotheses (Pintrich, 2002). Al-exander et al. (1998) also mention that the philosophical roots that underpin most empirical research are not identified or well described. It is therefore important to address the question how these differences in world hypotheses are reflected in the explanations of those labels. The main objective in this article is to facilitate this process by providing a dimensional framework to assess some of those paradigmatic differences.

Background and research questionReviewing contemporary epistemological research in search for the differ-

On the nature of epistemologies

41

ences and commonalities between the different labels, we found that many researchers fall short explaining these in great detail especially to the world hypotheses that are implicated. It could well be the case that researchers do have more in depth ideas and views about the construct definition but journal format regulations prohibit this. Moreover, it is also thinkable that researchers for pragmatic reasons choose not to include this kind of information because they want to publish in a certain journal. Finally, it could also be that this ex-ercise would confront researchers with insights they themselves have not yet thought about and which would perhaps destabilize their former ideas.

Although we hardly found thorough explanations, we did however come across many different instruments (formats) ranging from open-ended inter-views, the use of vignettes to the use of Likert-type questionnaires and the use of observations. Having no explicit account to the labels, we wondered whether the design of the research instrument could indirectly reveal informa-tion about the nature (referring to the world hypotheses assumptions) of the label. This indirect linkage is visualised in Figure 1. We assume that sound research characterises itself by some kind of congruence in the phases of a research. Every researcher, consciously or not, is led by world hypotheses (Pintrich, 2002). We content that these world hypotheses are reflected in the description of the primary research concepts and also parallel the research instrument (design) that seems most appropriate to investigate the prime concepts.

FIGURE 1. Linkages between world view and choice of research instruments

In line with this congruence assumption, we claim (see Figure 1 – infer-

Worldview

Concept definition

Choice instrument

INFE

RE

NC

E

CO

NG

RU

EN

CE


42

ence) that our focus does not to be per se on the search for an explicit explana-tion, but that it is also possible to direct our focus to the researchers’ choice of instrument. The main research question became therefore:

How does a researcher’s choice of instruments implicitly reveal cues about how the researchers perceives the nature of the epistemological construct?

Following Schoenfeld (2001), referring to ‘the nature of’ we more specifi-cally mean:

a) What form do epistemologies take?b) How are epistemologies organized?c) How are epistemologies accessed?

METHOD

A first step in collecting literature and instruments on epistemologies was to identify relevant articles in the scientific literature. The Social Science Citation Index was used for this aim. Next, we collected as many instruments as we could. Sometimes these instruments were published completely in the articles in which they were reported. In other cases, we sent an email to the authors of an article, often developers of the instruments, to ask for the complete version. In this way we collected 30 instruments on personal epistemology.

Table 1 provides an overview of some of the instruments that surfaced over the years; labelled by the authors and further categorised by subject matter domain. It must be said that it is not our contention to be exhaustive, i.e. it is hardly possible to obtain all available instruments for several reasons. First, researchers often do not label their research as explicitly “measuring epistemologies”. Secondly, the domain of research to personal epistemol-ogy is comprehensive, meaning that it is done within several subject matter domains. To oversee these domains totally requires that one is expert to all these domains. Thirdly, although it may be hard to obtain all the instruments, it also seems not necessary because many instruments are somewhat revised versions of previous developed ones.

The criterion for including an instrument was the occurrence of the term epistemology and / or a derivative thereof. The information about the instru-ments was gathered by studying the instruments themselves and in cases we could not get hold of a copy by studying the relevant articles.


43

Source Definition Outcomes Format Ori/deri

General

1. Perry (1968,1988) bo epbel epista inview ori

2. King & Kitchener (1994, 2002) bo/art epbel epidi inview pa

3. Kitchener & et al. (1999) art epbel epidi cofor pa

4. Boyes & Chandler (1990, 1991) art epbel epista inview insp

5. Baxter Magolda (1992, 1994, 1996, 2002) art epbel epilev oeq simto

6. Belenky et al. (1986) & Mc Vicker Clincky (2002) bo/art epbel epicat inview aot

7. Schommer (1990, 1994 a & b, 2002) art epbel epidi liquest insp

8. Jehng et al. (1993) art epbel epidi liquest pa

9. Schraw et al. (2001) art epbel epidi liquest pa

10. Kuhn et al. (2000, 2001, 2002) art epbel epipo quest (fc) nk (nfv)

11. Hofer (1997, 2000, 2001 a & b, 2002) art epbel epidi cofor pa (nfv)

12. Roozendaal et al. art epbel epidi quest pa (nfv)

13. Tirri et al. (1999) art sedef/epica epistan inview oe/c insp

14. Lonka et al. (1999) art epist epifac liquest

15. Lyons (1990) art epipers epipers castud pa

16. Philips (1998, 2001) art epibel epidi liquest simto

Mathematics

17. De Corte et al. (2002) art epbel epifac liquest pa

18. Roulet (1998) the epbel epistanc castud/cofor pa

19. Schoenfelt (1983, 2000 a & b) art epbel epibun castud/cofor ori

20. Hill (1993, 1997, 2000) art epbel epista quest/inview insp

21. Speer & Aguirre (2000) art epbel epibun cofor pa

Physics

22. Hammer & Elby (2001, 2002) art epiriso cofor insp

23. DiSessa (1993) art epiriso p-prims cofor ori

Science

24. (Butler) Songer & Linn (1990, 1991) art voS/sedef voS cofor shoq ori

25. Yerrick et al. (1998) art sedef epistanc cofor

26. Benson (1989) art vok epiview (2/3?) castud/cofor ori

27. Lederman (1987) art epicat epifac cofor

28. Hashweh (1996) art epibel quest

29. Maor et al. (1995) art teaepi teaepi (2) cofor ori

30. Abd-el-khalick et al. (1998) art nos/epicom cofor

TABLE 1. Instruments on epistemologies and some descriptive characteristics


44

Explanations of abbreviations in Table 1: Sources: comprises the type of sources that are conducted: article (=art); thesis (=the); book (= bo); Definition: construct definition: epistemologic belief (=epbel); epis-temological development (=epidel); epistemological cognition (= epicog); reflective judgement (=rejud); epistemic doubt (=epidob); epistemological reflection (=epiref); epistemological categories (=epicat); epistemologi-cal perspectives (=epipers); epistemological style (=episty); epistemologies (=epist); several definitions (=sedef); Outcomes: The terms in which the outcomes of the assessment are stated: epistemological categories (=epi-cat); epistemological levels of reflective judgment (=epilev); epistemologi-cal stages (=epista); epistemological stances (= epistanc) epistemological dimensions (=epidi); epistemological factors (= epifac) epistemological positions (=epipo);epistemological perspectives (=epipers); epistemologi-cal standards (=epistan); Format: Likert-questionnaire (=liquest); question-naire (=quest) case study (=castud); interviews (inview); cobined formats (=cofor); open-ended questions (=oeq); open-ended (=oe); closed (=c); semi-structured (=ss); Original/derival (=ori/deri): original (=ori) inspired (=insp); partly adopted (=pa); wholly adopted (=wa); similar to (=simto); as opposed to (=aat)

Next, in an iterative manner, analysing the instruments and studying the scientific articles on epistemologies (Chinn, 1993; diSessa, 1993; Duell & Schommer-Aikins, 2001; Elby & Hammer, 2001; Hofer & Pintrich, 1997; Hofer & Pintrich, 2002) and cognitive psychology (Davis & Sumara, 1997, 2001, 2002a, 2002b; Davis, Sumara, & Kieren, 1996; Kincheloe & Steinberg, 1993; Säljö, 2002; Shapiro & Kirby, 1998; Sumara & Davis, 1997; Varela, Thompson, & Rosch, 1997; Widdershoven, 1999), we developed a six-dimen-sional framework; a framework that would enable us to make explicit some of the assumptions about the nature (regarding the organization, form and access) of the construct underlying the design of the instrument. For reasons of trustworthiness and viability of the results, the framework was discussed amongst colleagues (member check: Lincoln & Guba, 1985) and presented to international educational research forums (auditing: Schwandt & Halpern, 1988): (1) the meeting of the Special Interest Group on Learning and Profes-sional Development of the European Association of Research on Learning and Instruction (EARLI) in Turku, Finland, 2000; (2) the Junior Researchers


45

(JURE) conference of EARLI in Amsterdam, The Netherlands, 2001; and (3) the ICO Summer School in Rethimo, Greece, in 2002.

RESULTS

In all, we developed a framework consisting of six dimensions. It enables a researcher to verify whether one’s espoused assumptions about the organi-sation, form and access of epistemologies are congruent with the assump-tions that underlie his or her choice of instrument and thus the design. For example, dimension five represents the issue whether it is possible to speak explicitly about epistemologies.

If a researcher holds no particular opinion about this matter, but decides to use a questionnaire, he or she implicitly takes a stand to this issue by choos-ing this particular instrument. This also holds for the other dimensions. Di-mension one raises the question whether epistemological beliefs consist of different epistemological factors or whether it is impossible to measure separate epistemological belief components.

Dimension two refers to the question whether beliefs are static or more malleable in nature. Dimension three refers to the issue whether epistemolo-gies are cognitive phenomena meaning solely existing in the heads of people or that epistemologies are also visible as non-cognitive structures in our world. Di-mension four asks if epistemological beliefs are pitched on a general level or perhaps are more contexts specific. Moreover, how researchers think about and define context also differs greatly (Pintrich, 2002).

Finally, dimension six calls into question whether it is feasible to speak openly about one’s epistemological beliefs.

The six dimensionsFigure 2 graphically represents the six dimensions. However, in order to

be able to scale the instruments along the different dimensions a more com-prehensive explanation is necessary. In the subsequent paragraphs we will explain these dimensions in more detail. The fist two dimensions take up the organization that is assumed through the researchers’ choice of an instru-ment.


46

FIGURE 2. A framework of six dimensions for analysing the nature of epistemologies

Isolated vs. EntwinedMany researchers use (Likert-type) questionnaires to elicit epistemologies

(Duell & Schommer-Aikins, 2001). These researchers assume it is possible to identify different isolated epistemological dimensions as measured with sub-sequent confirmatory and / or exploratory factor analyses. On the other end of the continuum there are researchers that question whether epistemologies can be dissected into different belief components at all. Using more natu-ralistic or ethnographic research instruments like open ended, non-directive interviews and/or participant observation, is seems hard to discern separate epistemological factors.

Researchers in the latter sense assume the whole composes more than its individual parts. Pintrich (2002) stresses that those different authors, at the end, call upon different incommensurable theories.

“In this manner the difference between these cognitive developmental models and the more cognitive perspective,…, reflects the traditional organismic and mechanistic metathe-ories that underlie the models. That is, the cognitive developmental models assume that epistemological thinking is more qualitative , organic, or holistic (the whole is greater than the parts) and can’t be broken down into independent components, at least not coherently. In contrast the more mechanistic cognitive models do assume that there can be a quantita-

How are epistemologies organised?

What form do epistemologies take?

How are epistemologies accessed?

1. Isolated

2. Static

3. Cognitive

4. General

5. Explicit

6. Self referential

Entwined

Dynamic

Physical

Context-specific

Implicit

Attributing


47

tive analysis of the components and that these somewhat independent components can be described and summarized in some manner to represent the nature of epistemological think-ing.” (Pintrich, 2002)

Static vs. DynamicMany researchers refer to the label they use in terms of having it as some

sort of characteristic – static and hard to change (like for example in people having epistemological beliefs). The instruments appropriated and the language used signify this issue: questionnaires and the use of vignettes emphasize the static and stable nature of epistemologies.

Other researchers, focusing on the interface between people, point to the fact that epistemologies are co-created in ongoing actual classroom action (interaction). Epistemologies are thus emergent and therefore more malle-able and dynamic in nature (Freeman, 1991, 1996a, 1996b). Epistemologies in this latter sense do not solely exist as unchanging, static cognitive compo-nents nor are they solely the result of dialogical processes, as exemplified by the Soviet psychologists like Bakhtin. According to Freeman both processes are taking place:

“However, a central question remains about the conceptions of practice which the teach-ers come to articulate and act upon through their experience in the in-service program. Do these conceptions exist a priori, but on a tacit level as Shulman implies, to be “made ex-plicit” through the process of teacher education? Or are they formed through the participa-tion in professional training? This study would suggest that both processes are taking place simultaneously and interactively… So the process of making the “the tacit explicit” is not a linear one of revealing what is known; it seems to be a dialectical one in which familiar and tacit knowledge interacts with – and is reshaped by – newly explicit understandings.” (Freeman, 1991, p. 453)

According to Freeman (1991) epistemologies do have static components, which however, have to obtain their actual meaning and interpretation in ongoing practice. Epistemologies in the latter sense are recreated and co-cre-ated in interaction. This process of representational recreation, as Karmiloff - Smith (from Wiliam, 1998) calls it, is a process whereby individuals through repeating dialogue become aware and are more able to discuss their represen-tations (beliefs – our adding) on a more and more explicit level.


48

How are epistemologies formed?The next two dimensions focus on cues about the form epistemological

beliefs take and the formation process along which epistemologies are struc-tured.

Psychological vs. PhysicalSome researchers, focusing on individuals through self reporting meas-

urement techniques, assume epistemologies reside primarily within the in-dividual (in the heads of people as cognitive structures or components). Others researchers focus more on the ‘context’ on which epistemologies are inscribed:

“The values and beliefs of the dominant culture within society influence the ways that school systems are organized and how school buildings are built and serve to establish the rules and standards of behaviour within those settings. These values and beliefs are embod-ied in the artifactual world of the school setting. From the placement of the teacher’s desk on a platform in front of a room to the honouring of a picture of Einstein in a special place on the wall, values and belief are material in the ways that we organize the spaces and features of the classroom.” (Shapiro & Kirby, 1998)

Researcher in the latter sense point out to the fact we live in an already interpreted reality. This means that – although people perceive and interpret the world individually – physical structures (like for example the way a class is organised) surrounding us are subsumed by epistemological inscriptions or beliefs which set limits to the range of possible interpretations (Packer & Winne, 1995; Shapiro & Kirby, 1998; Roth, 1998). To elicit these epistemolo-gies, semiotic analyses and research methods are necessary (Hofer & Pintrich, 1997; Hofer & Pintrich, 2002; Lakoff & Johnson, 1999; Shapiro & Kirby, 1998).

General vs. Context-specificSome researchers use more ethnographic methods (like for example de-

tailed observations of class discussions) to adhere epistemologies. In this way they point to the fact that epistemologies should be seen as context specific phenomena. Other researchers using questionnaires with general items (for


49

example: “nothing is certain but death and taxes”) assume epistemologies function on a more general level. Moreover, the issue of context-dependency of epistemologies is further hampered by the discussion how to define con-text-dependent. According to Pintrich (2002) the issue eventually is pitched on differences about the underlying meta-theory:

“The more important issue concerns the refinement of our definitions of domains and contexts. For example, for many of the models in this volume, particularly the more de-velopmental ones, domains refer to larger areas of our lives, from academic to work to the personal and social arenas of our lives. For other more cognitive models, domains have been synonymous with school subject areas or disciplines. Finally, for strong contextualist models, domains may be to broad a category and these models often propose a focus on epistemologi-cal thinking within a domain, but also how the classroom context represents and shapes the nature of knowledge and knowing within that domain. Accordingly, there is a need for clear statements about domain boundaries in our models.” (Pintrich, 2002, p. 399)

Generally speaking, one can safely state that ‘contextualists’ tend towards the use of more ethnographic, naturalistic methods in which it is not so much their purpose to get attached from the context particularities to more abstract generalisable findings, but giving detailed insights into, and be as sensitive as possible to, the context from which epistemologies emerge. The research methods used by researchers in the latter sense are more narrative and not so much focusing on prediction but understanding.

How to access and elicit epistemologies?The access to epistemologies is the final question when asking about the

nature of epistemologies. The first dimension under this heading refers to the question if it is possible to be explicit about ones epistemologies. The second dimension refers to the question whether it is wise to be direct in asking peo-ple about their epistemologies.

Explicit vs. ImplicitThe study of epistemologies generally focuses on an analyses of what

people say they believe. Words are taken as providing a vehicle for thought, and people are ‘taken at their word’. To ‘take teachers at their word’ means to believe in the capacity and willingness of people to talk about themselves


50

in epistemological terms. In case of teachers epistemologies, teachers’ words are thus taken for their capacity to reveal what is in the teachers’ minds and therefore to represent their beliefs and thinking.

The dimension Explicit vs. Implicit in this respect refers to the question whether researchers are positive about the respondents’ capacity to reflect on their epistemological beliefs explicitly and therefore ‘take them at their word’, or whether this is impossible because teachers do not have the language or the willingness to discuss epistemological matters openly and directly. In the latter example some researchers do not ask teachers directly and explicitly about their epistemological beliefs but infer these for example through de-tailed language analyses (Siskin, 1994).

Self – referential vs. AttributingThe last dimension Self-referential vs. Attributed points to the issue that even

in the case researchers are positive about the respondents ability to explicitly and directly discuss epistemological beliefs, according to some researchers one must take into account that what they profess to believe and what they actually believe may be two different things. This becomes clear when look-ing at actual school practice. What people profess to believe may not be trans-lated and visible in straightforward ways. Some researchers think therefore it’s wiser not to take teachers at their word but to focus on beliefs one is able to attribute, based on detailed analyses of behaviour and natural conversations.

“A growing literature on teachers beliefs makes it clear that what teachers profess to believe and what they actually believe may be very different things; moreover, that what they ‘believe’ may not translate into classroom practice in straightforward ways. Just as in all other domains, beliefs shape the goals teachers have for classroom interactions. They shape the set of options teachers believe to be available to meet those goals, and they shape the ways in which those resources are employed. A comprehensive model of teaching must delineate the nature of teachers beliefs and elaborate the ways in which they work.” (Schoenfeld, Minstrell & Zee, 2000)

The instruments scalingIn the previous section we explained the different dimensions in more

detail. This explanation is necessary to be able to scale and weigh the differ-


51

ent instruments against each of the dimensions. In the figures 3 to 8 four instruments – as an example – are scaled to each of the six dimensions. Over the years numerous domain specific and domain general (see Table 1) epistemological belief instruments were developed. Given the limited space, the instruments which are selected and provided in this article are chosen for their diversity in the formats which they represent. First a short verbal de-scription of each of these four instruments is provided. With this description the reader is able to infer and verify our ranking of the instruments to the six dimensions. Although the scaling process is done with great care, the reader should keep in mind that this framework first and foremost has to draw the discussion to the assumptions regarding the organisation, form and accessi-bility to a more explicit level. This discussion should not get bogged down to a dispute whether our scalings are precisely the right ones.

W.G. Perry (P) – (Perry, 1968, 1988)In the fifties and sixties, William Perry, Jr., conducted open-ended inter-

views with undergraduates at the end of each year, opening with the ques-tion: ‘What stands out for you this year?’ In this way he hoped to capture the students’ own perceptions of their learning and development, not so much in terms of academic achievement as in overall changes they experienced within themselves. He identified that students’ ways of perceiving knowledge and values changed during the college years. Epistemological beliefs – as identi-fied by Perry using non-directive, open interviews – are like implicit – mean-ing that the student himself is not so much consciously aware of – filters through which students evaluate and encounter the challenges they meet.

M. Schommer-Aikens (SA) – (Duell & Schommer-Aikins, 2001; Schommer, 1990a; 1994a; 1998)

In the nineties, Marlene Schommer-Aikens, was the first who reported a major study to test the assumption of epistemological beliefs as a system of more of less independent beliefs. Independent beliefs are beliefs that do not necessarily have to mature congruently and simultaneously. By referring to a system of beliefs she meant that there is more than one belief dimension to


52

consider in personal epistemology. In order to assess these beliefs a Likert-type questionnaire was developed. Students had to respond to statements such as: ‘Scientists can ultimately get to the truth.’ or ‘I don’t like movies that have an open ending.’

N. Lyons (L) – (Lyons, 1990, 1995) Lyon speaks of ‘nested epistemologies’ to point to the fact that teachers

and learners have interconnected and interacting epistemological perspec-tives. This means that values and ideas, teachers’ subject knowledge, their relationships with students and conceptions about themselves and of their students as epistemological subjects are part of and create a certain predispo-sition towards knowledge definitions. Lyons uses non-directive interviews as her major way of collecting data. Excerpts from these interviews are used to exemplify the credibility in which she uses the term epistemological beliefs.

J. Aguirre & N. Speer (A&S) – (Aguirre & Speer, 2000; Schoenfeld et al., 2000)The method Aguirre and Speer (2001) propose is using detailed videotapes

and audiotapes of actual classroom algebra lesson conversations. The parsing of this information is phenomenological. The analysis is often carried out in a line-by-line level seeking how and why teachers make decisions and the influence of certain bundles of beliefs on those decisions. Epistemological beliefs – although not explicitly stated – are part of these beliefs bundles.

FIGURE 3. Dimension I – Regarding the organization: ISOLATED VERSUS ENTWINED

ISOLATED ENTWINED

PSA

A&S L


53

Explanation: Given the vignettes of these four researchers and their in-struments, we identify SA on the left side of the continuum because her main research method – a Likert type questionnaire – couches the implicit assump-tion that epistemologies are organized into different epistemological factors or dimensions which can be identified through factor analyses. L on the other hand using fragments from interviews with teachers and their students organ-izes epistemological beliefs as holistic features which cannot be dissected into mutually exclusive factors. The ones in between from P and A&S presume an organization – using respectively open ended questionnaires and observations – that is more holistic than fragmented or isolated

FIGURE 4. Dimension II – Regarding the organization: STATIC VERSUS DYNAMIC

Explanation: To the dimension static versus dynamic Lyons without question has the most dynamic meaning malleable definition to the episte-mological perspective.

Using interview transcripts the triggering of epistemologies is dependent on the context one is in. A rather situated perspective is also visible within the research of A&S. In their case epistemological beliefs elicitation also depends on observations.

SA on the other hand using questionnaires assumes epistemological be-liefs are not so malleable and situational but instead are organized as theories or traits. Perry – finally – organizes epistemological beliefs as stage like phases. Students change their epistemological outlooks as college continuums. How-ever they are not as malleable as the perspectives from L and A&S

STATIC DYNAMIC

SAP

A&SL


54

FIGURE 5. Dimension III – Regarding the form: COGNITIVE VERSUS PHYSICAL

Explanation: SA,P,S applying instruments that focus on the individual sense making adopt implicitly a cognitive psychological perspective that as-sumes epistemologies to reside solely in the heads of people thus depicting them as cognitive constructs.

L on the other hand focusing on the interface between people leaves open the issue whether beliefs could also exist as and surpass the rigid boundaries of cognitive psychology. That’s why we do not depict her.

FIGURE 6. Dimension IV – Regarding the form: GENERAL VERSUS CONTEXT-SPECIFIC

Explanation: The dimension general versus context specific refers to the extend the instrument is formed around the actual practice of students or teachers.

In case of the questionnaire that is used by S, sample items are stated very generally “Nothing is certain but death and taxes.” and “I don’t like movies that have an open ending.” The model by P denotes a different general epis-

COGNITIVE PHYSICAL

LSA, P, S

GENERAL CONTEXTUAL

PSA

A&S L


55

temological outlook which also is not situation specific. The instruments by A&S and L are formed around class encounters between student and teachers and are thus more attuned to the actual learning / teaching context.

FIGURE 7. Dimension V – Regarding the access: EXPLICIT VERSUS IMPLICIT

Explanation: To what extend do the instruments provide the respond-ents with the capacity to talk about their epistemological beliefs explicitly. In case of SA her questionnaire is developed in a manner that presumes that students quite directly have access to epistemological issues. In case of L stu-dents and their teachers are not so epistemologically ‘literate’.

This is because epistemologies are often unconscious and hold intricate relations with issues of self. P characterises students’ epistemologies talking about what stood out for them in a school year. A&S don’t think it is neither wise nor possible to have access to epistemologies explicitly because people do not have the language to talk about these issues.

FIGURE 8. Dimension VI – Regarding the access: SELF REFERENTIAL VERSUS ATTRIBUTING

EXPLICIT IMPLICIT

PL

SAA&S

SELF REFERENTIAL

ATTRIBUTING

P

LSA

A&S


56

Explanation: This dimension is a further normative refinement of the dimension V. It questions whether respondents are able and moreover if it is feasible to talk about their epistemological beliefs. SA as do also the cognitive perspectives from P and L think it is possible and prudent to let respondents explicitly report their epistemologies. A&S are the exceptions to this dimen-sion. They think in case of self reporting measures respondents fall in danger of social desirability. It is therefore – more reliable – to attribute epistemolo-gies to people.

DISCUSSION

Reviewing the domain at personal epistemology, we encountered a re-search area that is characterised by many disparate labels (Hofer & Pintrich, 2002; Hofer, 2000). Moreover, these labels are grounded in and supported by different paradigmatic communities from which they spring (Pintrich, 2002). To the clarity and further development of the field it is imperative that those differences be explicated on a more explicit level. Since a direct comparison of these labels is not possible because the labels are not explained in great detail, we revealed these assumptions, taking the choice of instrument as our starting point. To take this position was possible given the default assump-tions taken in this article, that research throughout its research phases (from the worldview explanation to the choice of instrument) is congruent. We thus questioned whether the design of an instrument could tell us something about the assumptions that characterise the nature of epistemologies mean-ing its form, organization and access (see Figure 2).

In an iterative manner, analysing the instruments themselves and the liter-ature on epistemologies and cognitive psychology, we developed a framework for these assumptions. This framework enables researchers to make explicit, or if already available, to consciously reflect upon, this choice and place the assumptions against the dimensions developed in our framework. For ex-ample: Researcher A talking about epistemological beliefs uses a Likert-type questionnaire. Does he however agree with the assumptions that are impli-cated by this design? He may or may not. The issue of importance is that the fundamental discussion to the nature of these constructs is facilitated and drawn to a more explicit level.


57

However, having developed a framework to facilitate and categorise as-sumptions about the nature of epistemologies by looking at the researchers choice of instrument does in itself not supply the researcher with a thorough operationalisation of the label he or she chooses to apply.

Referring to Figure 2 we contend that any careful and thorough operation-alisation should take into account and be grounded in an explicit worldview. We content that in every research a worldview is assumed – perhaps uncon-sciously but never coincidently – that guides the researcher in his or her justi-fication and explanation of the research construct (Jung, 1981).

The framework developed in this article allows researchers to take a first step explicating those fundamental worldview assumptions. Kagan (1990) also addresses the issue posed in this article that different techniques point to different world hypotheses but also that it is hard for researchers to judge the validity of a technique unless he or she is familiar with the relevant founda-tions.

“After reviewing various approaches to the assessment of teacher cognition, it becomes clear that one deeply problematic aspect of this literature is that techniques are derived from different epistemological traditions, Likert scales are derived from psychometric test theory, textual analyses of teachers’ language from theory relating language to cognition, self-reflec-tion hierarchies from studies of cognitive mediation, and concept mapping from schema theory and research on memory for verbal information. Each of these traditions provides a different definition of evidence and truth; some contradict each other. It is difficult for a reader to judge the validity of a technique unless he or she is familiar with the relevant foundation, and few consumers (non-specialists) could be expected to have such an eclectic background.” (Kagan, 1990, p. 451)

Kagan is quite pessimistic about the ability of a researcher to transcend his or her point of view because one can’t be an expert to all foundations. We agree, although it seems to us that it is not just a question to the ability of a researcher to transcend his or her foundation but also the willingness of researchers to be open and susceptible to other world hypotheses:

“I have found over and over again that the acceptance of a new point of view…has

much less to do with the validity of that point of view than with (one’s) readiness to consider any alternatives whatsoever.” (Schein, 1987, p. 107 – from Evans, 1996)

To be genuinely interested and knowledgeable about the differences in


58

terms of underlying worldviews requires courage; Courage to take ones most innate and taken for granted assumption to a critical test. Referring to re-cent literature on epistemologies we notice that new stances get serious and considerable attention (Hammer & Elby, 2002; Hofer & Pintrich, 2002). We welcome this attention and openness and hope that in combination with a renewed discussion to the foundations of the domain the field will prosper and become ready for the coming century.

REFERENCES

Abd-el-Khalick, F. (1998). The nature of science and instructional practice: making the unnatural natural. Journal of Research in Science Teaching.

Aguirre, J., & Speer, N. M. (2000). Examining the relationships between be-liefs and goals in teacher practice. Journal of Mathematical Behavior, 18, 327-356.

Baxter Magolda, M. B. (1992). Student’s epistemologies and academic ex-periences: implications for pedagogy. The Review of Higher Education, 15, 265-287.

Baxter Magolda, M. B. (1994). Post-college experiences and epistemology. The Review of Higher Education, 18, 25-44.

Baxter Magolda, M. B. (1996). Epistemological development in grade and professional education. The Review of Higher Education, 19, 283-304.

Baxter Magolda, M. B. (2002). Epistemological reflection: The evolution of epistemological assumptions from age 18 to 30. In B. K. Hofer & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 89-102). Mahwah: Lawrence Erlbaum Associates.

Belenky, M. F., Clinchy, B. M., Goldberger, N. R., & Tarule, J. M. (1986). Women’s ways of knowing: The development of self, voice and mind. New York: Basic Books.

Benson, G. D. (1989). Epistemology and science curriculum. Journal of Cur-riculum Studies, 21, 329-344.

Boyes, M., C., & Chandler, M. (1991). Cognitive development, epistemic doubt, and identity formation in adolescence. Journal of Youth and Adoles-cense, 21, 277-304.

Bruner, J. (2002). Making stories: Law, literature, life. New York: Farrar, Straus


59

and Giroux.Butler Songer, N., & Linn, M. C. (1991). How students’ views of science

influence knowledge integration? Journal of Research in Science Teaching, 28, 761-784.

Chandler, M., & Boyes, M., C. (1990). Relativism and stations of epistemic doubt. Journal of Experimental Child Psychology, 50, 370-395.

Chinn, C. A., Brewer, W. F. (1993). The role of anomlous data in knowledge acquisition: a theoretical framework and implication for science instruc-tion. Review of Educational Research, 63, 1-49.

Davis, A. B., & Sumara, D., J. (1997). Cognition, complexity, and teacher edu-cation. Harvard Educational Review, 67, 105-125.

Davis, A. B., & Sumara, D., J. (2001). Learning communities: Understand-ing the workplace as a complex system. New Directions for Adult Continuing Education, 92, 85-95.

Davis, A. B., & Sumara, D., J. (2002a). Constructivist discourses and the field of education: problems and possibilities. Educational Theory, 52, 409-428.

Davis, A. B., & Sumara, D., J. (2002b). Constructivist discourses and the field of education: problems and possibilities. Educational Theory, 52, 409-428.

Davis, A. B., Sumara, D., J., & Kieren, T. (1996). Cognition, co-emergence, curriculum. Journal of Curriculum Studies, 28, 151-169.

De Corte, E., Eynde, P. O. t., & Verschaffel, L. (2002). “Knowing what to be-lieve”: The relevance of students’ mathematical beliefs. In B. K. Hofer & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowl-edge and knowing (pp. 197-320). Mahwah: Lawrence Erlbaum Associates.

diSessa, A., A. (1993). Toward an epistemology of physics. Cognition and In-struction, 10, 105-225.

Duell, O., K., & Schommer-Aikins, M. (2001). Measures of people’s beliefs about knowledge and learning. Educational Psychology Review, 13, 419-449.

Elby, A., & Hammer, D. (2001). On the substance of a sophisticated episte-mology. Science Education, 85, 554-567.

Evans, R. (1996). The human side of school change: Reform, resistance, and the real-life problems of innovation. San Francisco: Jossey-Bass Publishers.

Firestone, W. A. (1987). Meaning in method: The rhetoric of quantitative and qualitative research. Educational Researcher, 16-21.

Freeman, D. (1991). “To make the tacit explicit”: Teacher education, emerg-ing discourse, and conceptions of teaching. Teaching & Teacher Education,


60

7, 439-454.Freeman, D. (1996a). “To take them at their word”: Language data in the

study of teachers’ knowledge. Harvard Educational Review, 66, 732-761.Freeman, D. (1996b). The ‘unstudied problem’: Research on teacher learning

in language teaching. In D. Freeman & J. C. Richards (Eds.), Teacher learning in language teaching (pp. 351-378). Cambridge: Cambridge University Press.

Hammer, D., & Elby, A. (2002). On the form of a personal epistemology. In B. Hofer, K. & P. R. Pintrich (Eds.), Personal epistemology (pp. 169-190). Mahwah, New Jersey: Lawrence Erlbaum Associates.

Hashweh, M. Z. (1996). Effects of science teachers’ epistemological beliefs in teaching. Journal of Research in Science Teaching, 33, 47-63.

Hofer, B., K. (2001). Personal epistemology research: Implications for learn-ing and teaching. Journal of Educational Psychology Review, 13, 353-383.

Hofer, B., K., & Burr, J. E. (2001). The development of personal epistemology: The evolving coordination of the subjective and objective. Paper presented at the EAR-LI, Fribourg, Switzerland.

Hofer, B. K. (2000). Dimensionality and disciplinary differences in personal epistemology. Contemporary Educational Psychology, 25, 378-405.

Hofer, B. K., & Pintrich, P. R. (1997). The development of epistemologcal theories: beliefs about knowledge and knowing and their relation to learn-ing. Review of Educational Research, 67, 88-140.

Hofer, B. K., & Pintrich, P. R. (2002). Personal epistemology: The psychology of be-liefs about knowledge and knowing. Mahwah, New Jersey: Lawrence Erlbaum.

Jehng, J. J., Johnson, S. D., & Anderson, R. C. (1993). Schooling and students’ epistemological beliefs about learning. Contemporary Educational Psychology, 18, 23-35.

Jung, C. G. (1981). The structure and dynamics of the psyche. Princeton, NJ: Prin-ceton University Press.

Kincheloe, J., L., & Steinberg, S., R. (1993). A tentative description of post-formal thinking: The critical confrontation with cognitice theory. Harvard Educational Review, 63, 296-320.

King, P. M., & Kitchener, K. S. (1994). Developing reflective judgement: Understand-ing and promoting reflective growth and critical thinking in adolescents and adults. San Francisco: Jossey-Bass.

Kitchener, K. S., Wood, P. K., & Jensen, L. (1999 August). Curricular, co-cur-ricular, and institutional influence on real-world problem solving. Paper presented at


61

the Annual meeting of the American Psychological Association, Boston, Massachussets.

Kuhn, D. (2001). How do people know? Psychological Science, 12, 1-8.Kuhn, D., Cheney, R., & Weinstock, M. (2000). The development of episte-

mological understanding. Cognitive Development, 15, 309-328.Kuhn, D., & Weinstock, M. (2002). What is epistemological thinking and why

does it matter. In B. K. Hofer & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 119-120). Mahwah: Lawrence Erlbaum Associates.

Lakoff, G., & Johnson, M. (1999). Leven in metaforen. Nijmegen: Uitgeverij SUN.

Lederman, N. G. (1987). Science teachers’ conceptions of the nature of sci-ence: Do they really influence teaching behavior? Science Education, 71, 721-734.

Lincoln, Y., S., & Guba, E., G. (1985). Naturalistic inquiry. Beverly Hills: CA: Sage.

Lonka, K., & Lindblom-Ylänne, S. (1999). Epistemologies, Conceptions of Learn-ing, and Study Success in Medicine and Psychology. Helsinki: Faculty of Medi-cine, University of Helsinki.

Lyons, N. (1990). Dilemmas of knowing: Ethical and epistemological dimen-sions of teachers’ work and development. Harvard Educational Review, 60, 159-180.

Lyons, N. (1995). Re-imagining teacher education. Curriculum Inquiry, 25, 75-85.

Maor, D., & Taylor, P., Charles. (1995). Teacher epistemology and scientific inquiry in computerized classroom environments. Journal of Research in Sci-ence Teaching, 32, 839-854.

Packer, M. J., & Winne, P. H. (1995). The place of cognition in explanations of teaching: a dialog of interpretive and cognitive approaches. Teaching and Teacher Education, 11, 1-21.

Perry, W. G. (1968). Forms of intellectual and ethical development in the college years. New York: Holt, Rinehart and Winston, Inc.

Perry, W. G. (1988). Different worlds in the same classroom. In P. Ramsden (Ed.), Improving learning: New perspectives (pp. 145-161). London: Kogan Page Ltd.

Phillips, F. (1998). Accounting students’ beliefs about knowledge: Associating


62

performance with underlying belief dimensions. Issues in Accounting Educa-tion, 13, 113-126.

Phillips, F. (2001). A research note on accounting students’ epistemological beliefs, study strategies, and unstructured problem-solving perfomance. Issues in Accounting Education, 16, 21-39.

Pintrich, P. R. (2002). Future challenges and directions for theory and re-search on personal epistemolgy. In B. K. Hofer & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 389-414). Hahwah: Lawrence Erlbaum Associates Publishers.

Roulet, G., R. (1998). Examplary mathematics teachers: subject conceptions and in-structional practices. University of Toronto, Toronto.

Säljö, R. (2002). My brain’s running slow today – the preference for “things ontologies” in research and everyday discourse on human thinking. Studies in Philosophy and Education, 21, 389-405.

Schoenfeld, A., H. (1983). Beyond the purely cognitive: Belief systems, social cognitions, and metacognitions as driving forces in intellectual perform-ance. Cognitive Science, 7, 329-363.

Schoenfeld, A., H. (2000). Models of the teaching process. Journal of Math-ematical Behavior, 18, 243-261.

Schoenfeld, A., H., Minstrell, J., & Zee, E. v. (2000). The detailed analyses of an established teacher’s non-traditional lesson. Journal of Mathematical Behavior, 18, 281-325.

Schommer, M. (1990a). Effects of beliefs about the nature of knowledge on comprehension. Journal of Educational Psychology, 82, 498-504.

Schommer, M. (1990b). Oustanding Dissertation Award 1989-1990: The ef-fect of beliefs about the nature of knowledge on comprehension. Reading Research Quarterly, XXV, 340-341.

Schommer, M. (1994a). An emerging conceptualization of epistemological beliefs and their role in learning. In R. Garner & P. A. Alexander (Eds.), Beliefs about text and instruction with text. Hillsdale, New Jersey: Lawrence Erlbaum.

Schommer, M. (1994b). Synthesizing epistemological belief research: tenta-tive understandings and provocative confusions. Educational Psychology Re-view, 6, 293-319.

Schommer, M. (1998). The influence of age and education on epistemologi-cal beliefs. British Journal of educational Psychology, 68, 551-562.


63

Schommer-Aikins, M. (2002). An evolving theoretical framework for an epis-temological belief system. In B. K. Hofer, Pintrich, P. R. (Ed.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 103-118). Mahwah: Lawrence Erlbaum Associates.

Schraw, G. (2001). Current themes and future directions in epistemological research: a commentary. Educational Psychology Review, 13, 451-464.

Schwandt, T., A., & Halpern, E., S. (1988). Linking auditing and metaevaluation. Newbury Park: Sage publications.

Shapiro, B., & Kirby, D. (1998). An approach to consider messages of school science learning culture. Journal of Science Teacher Education, 9, 221-240.

Siskin, L. S. (1994). Realms of knowledge: Academic departments in secondary schools. New York: Falmer Press.

Sumara, D., J., & Davis, A. B. (1997). Enactivist theory and community learn-ing: toward a complexified understanding of action research. Educational Action Research, 5, 403-422.

Tirri, K., Husu, J., & Kansanen, P. (1999). The Epistomological Stance be-tween the Knower and the Known. Teaching &Teacher Education, 15, 911-922.

Varela, F., J., Thompson, E., & Rosch, E. (1997). The embodied mind – Cognitive science and human experience (Sixth printing ed.). Cambridge, London: The MIT Press.

Widdershoven, G. A. M. (1999). Cognitive psychology and hermeneutics: Two approaches to meaning and mental disorder. Philosophy, Psychiatry and Psychology, 6, 245-253.

Wiliam, D. (1998). Being mathematical versus mathematical be-ing: situated activity, interactive emergence and mathematical education. Paper presented at the 24th an-nual conference of the British Educational Research Association, Queens’ University of Belfast.

Yerrick, R. K., Pedersen, J. E., & Arnason, J. (1998). “We’re Just Spectators”: A Case Study of Science Teaching, Epistemology, and Classroom Man-agement. Science Education, 82, 619-648.


65

TEACHER-CHILD RELATIONSHIP:A MEETING OF MINDS

Flavia Lecciso*, Olga Liverta-Sempio**, Antonella Marchetti***

From a Vygotskian point of view, this study highlights the relevance of considering the teaching-learning processes as characterized by the interaction of the child with the Other, where the other is both culture and person. The aim of the work is to explore the representation of the teacher-child relationship from both partners’ perspective, in connection with different functions linked to it, from the relational-affective ones to the teaching-learning ones. Thirty-five teachers, randomly selected among those that during a week spend the most of the time with children, were given The Teacher Relationship Interview (TRI). Thirt-five children, aged from 6 to 10 years, were given the Separation Anxiety Test (SAT) (Family and School versions) and the Child Relationship Interview. Analyses showed significant positive associations between the representations of the relationship that the actors establish (TRI and Child Relationship Interview) and the quality of the attachment the child has developed with the teacher (SAT School). The results showed that the adult-child relationship represents a “meeting between minds”, that is, an intersubjective process where the understanding of both our own and other people`s emotions and thoughts could become a rich opportunity for learning, and a protective factor for the child as well as an enrichment for both partners.

“Then said a teacher, speak to us of teaching. And he said:No man can reveal to you aught but that which

already lies half asleep in the dawning of your knowledge.The teacher who walks in the shadow of the temple,

among his followers, gives not of his wisdom but rather of his faith and his lovingness.

If he is indeed wise he does not bid you enter the house of his wisdom, but rather leads you

to the threshold of your mind.” The Prophet, Gibran Kahlil Gibran, 1967

* Catholic University of the Sacred Heart, Piacenza and Brescia, Italy.** Catholic University of the Sacred Heart, Milano, Italy.*** Catholic University of the Sacred Heart, Brescia, Italy.

Flavia Lecciso, Olga Liverta-Sempio, Antonella Marchetti

66

INTRODUCTION

The words reported above express in poetry some of the main theoretical concepts supporting our paper which, from a socio-cultural point of view (Bruner, 1990, 1996), highlights the advantage of considering the teaching-learning processes as characterized by the interaction of the child with the Other, where the other is both culture and person (Liverta-Sempio & Mar-chetti, 1995, 2001a).

According to this perspective, “teaching what lies half asleep in the dawning of your knowledge” is strongly connected to the educative role played in the Zone of Proximal Development (Vygotsky, 1978) that is defined in opera-tive terms by the Brunerian concept of scaffolding (Bruner, 1986).

Moreover, we intend to underline the active role of both partners (the adult and the child) involved in such a process.

Focus will be placed, on the one hand, on the teacher’s capacity to recog-nize the “space” of potentiality of the child and to scaffold him/her activ-ity in this zone, trying to calibrate the help in accordance with the contexts and the abilities progressively acquired by the pupil (Wood, Bruner & Ross, 1976); on the other hand, on the child, as an active partner in the Zone of Proximal Development, “invites” and “calls” the adult to take part in his/her process of growth, gradually assuming bigger responsibility in the different learning situations.

“The teacher who gives not of his wisdom but rather of his faith and his lovingness” refers to the importance, underlined by recent approaches (Pianta, 1999), of the relational-affective dimension for the educational relationship. In this sense, the socio-cultural perspective is integrated and enriched through the consideration of the affective dimension.

The reference is to Bion’s (1962) and Winnicott’s (1965) psychoanalytic theorizations and to the more recent declination of the attachment theory, both in representational terms (Main, Kaplan, & Cassidy, 1985) and in a mul-tiple-caregiver perspective (Pianta, 1999; van Ijzendoorn et al., 1992).

According to the attachment theory, the representational perspective claims that the individual differences in attachment are not only related to behavioural patterns, but also to language and mental structures; moreover the research on multiple caregivers suggests the significance of the different affective partners (or figures), for example educators and teachers, involved

Teacher-child relationship: a meeting of minds

67

in a child’s development. These trends allow us to recognize the adult-child relationship not only as

a place of growth, but also as an important resource in negative or risk situ-ations. Finally “the wise teacher who leads to the threshold of your mind” evokes the recent field of studies about the Theory of Mind, a theoretical perspective that analyses the human ability to understand one’s own and other people`s mental states (intentions, emotions, beliefs, false beliefs, etc.) and to predict and explain people’s behaviour through these mental inferences (Premack & Woodroof, 1978; Wimmer & Perner, 1983).

After a first phase in which this area of development was considered as purely “cognitive”, more recently a great deal of studies have been con-ducted from contextual and socio-affective points of view, trying to identify the individual differences (Astington, 1996; Liverta-Sempio, Marchetti & Lecciso, in press; Marchetti & Massaro, 2002; Repacholi & Slaughter, 2003) and the relational factors (Cutting & Dunn, 1999; Dunn et al., 1991; Fon-agy, Redfern & Charman, 1997; Liverta-Sempio & Marchetti, 2001a; Meins, 1997; Meins et al., 2003; Perner, Ruffman & Leekam, 1994) involved in the theory of mind development. In particular, in recent years, some authors have analyzed the link between a child’s mental ability and his/her affective dimension pointing out a significant connection between the comprehen-sion of the mind and the affective quality of the caregiver-child bond, that is not only between mother and child (Fonagy & Target, 2001; Meins, 1997) but also between teacher and pupil (Lecciso, Liverta-Sempio & Marchetti, 2002; Liverta-Sempio & Marchetti, 2001a).

In this perspective, the security of the attachment relation results linked to a high capacity of mentalization by the child.

The explanation of such a link is related by the authors to the caregiver’s role. According to this aspect of the relationship, Elizabeth Meins (Meins, 1997; Meins et al, 2001; Meins et al., 2003; Meins & Fernyhough, 1999;) pro-poses the concept of Mind-Mindedness, that is the proclivity to treat the child as a different individual with a separate mind; Peter Fonagy (Fonagy, Redfern & Charman, 1997; Fonagy & Target, 2001) calls it the Reflexive Function, that is the caregiver’s capacity of containment, meant as a meta-cognitive competence, that does not only enable him/her to conceive the child as a mental subject, but also to reflect such an image to him/her.

In this perspective, the child with a secure attachment is not only able to


68

explore the physical and social word, but also his/her own mind as well as other people`s minds.

The concept that Fonagy and Target (2001) and Meins (1997) adopt to understand the mother-child relationship can be applied to the teacher-child relationship: from this point of view, the teacher’s mind can be a place which the pupil can explore. The aim of this work, through an integration of the above mentioned theoretical points of view, is to explore the representation of the teacher-child relationship viewed from both partners’ perspective. More precisely, our goals were:

• to analyse the representation that pupil and teacher have about their relationship;

• to analyse the representation that pupil and teacher have about their relationship and the quality of pupils’ attachment to parents and teach-er;

• to explore the relation between teacher’s and child’s representations of their relationship.

METHOD

ParticipantsParticipants were 35 teachers randomly selected among those that during

a week spend most of the time with children; and 35 children aged from 6 to 10. The children were from predominantly middle class backgrounds.

Procedure and MeasuresTeachers and children were tested at school. The teachers were given the Teacher Relationship Interview (Pianta et

al., 1999; Italian translation: Liverta-Sempio & Marchetti, 2001b). It is a semi-structured interview which elicits teachers’ representations of their relationship with a specific pupil. It contains questions about a teacher’s description of her1 relationship with that pupil, and about

1 All the teachers interviewed were women.


69

specific topics or themes/situations such as discipline, achievement and affect.

Teachers were probed throughout the interview to provide specific examples for their characterizations of the child, and for thoughts and feelings associated with these examples/episodes. In the scoring sys-tem the teachers’ mental representations of their relationship with a child were assessed according to 11 dimensions or scales grouped in 4 areas: content or themes represented; process or how the teacher rep-resents him/herself; affective tone of representations; domain specific representations (see table 1).

A 3-point scale (0= no evidence of construct; 1= minimal evidence of construct; 2= clear evidence) was adopted to score each response to a single question in the interview on scales in the first 3 areas; the domain codes were simply scored as present (1) or absent (0).

The children were given the following tests and interview: SAT-Family (Separation Anxiety Test) (Klagsbrun & Bowlby, 1976;

Fonagy, Redfern & Charman, 1997; Liverta-Sempio, Marchetti & Lecciso, 2001; Slough, Goyette & Greenberg, 1988) and SAT-School (Liverta-Sempio, Marchetti & Lecciso, 2001), which measures the at-tachment relationship of the child with the parents and the teacher, respectively.

They are semi-projective representational tests in which situations of child’s separation from his/her caregiver are described and presented to the child through photographs.

According to Slough, Goyette and Greenberg (1988), for each test there are three situations of mild separation (for example, in SAT-Family: mother tucks child up in bed and leaves room) and three sit-uations of severe separation (for example for SAT- School: teacher is going away for two weeks. He/she leaves the pupils with another teacher). Following the explanation of each situation, the examiner asks the child:

• How does ….(name of the tested child) feel?• Why does ….(name of the tested child) feel so?• What’s ….(name of the tested child) going to do?

The scoring system was the one devised by Slough, Goyette and


70

Greenberg (1988). The scoring indices and ratings were designed to reflect three dimensions of the children’s answers:

• Attachment dimension, that is, expression of vulnerability/need in the severe situations of separation;

• Self-reliance dimension, that is, expression of self-confidence in the situations of mild separation;

• Avoidance dimension or degree of elusion of the discourse about separation.

The three dimensions of attachment responses coded from the SAT were combined to achieve an overall SAT score (Fonagy, Redfern & Charman, 1997). This overall SAT score allowed to rate the subjects in three groups: group of secure attachment; group of insecure attach-ment; group of ambiguous attachment (neither secure, nor insecure).

The CHILD RELATIONSHIP INTERVIEW (IRBI in Italian: Inter-vista sulla Relazione del Bambino con l’Insegnante) (Liverta-Sempio, Marchetti & Lecciso, 2002). It is a semi-structured interview which analyses the children’s representations of their relationship with a spe-cific teacher. Children were asked to remember specific events about the relationship with that teacher, such as agreements, disagreements, as well as thoughts and feelings associated with these episodes. The scoring system was partially based on the scoring system developed by Pianta et al. (1999) for TRI. The childs’ mental representations of his/her relationship with a teacher were assessed according to 11 di-mensions or scales grouped in 4 areas: content or themes represented; process or how the child represents him/herself; affective tone of rep-resentations; domain specific representations (see table 1).

Each teacher was interviewed about the relationship with a specific pupil, who in turn was interviewed about the teacher who spoke about him/her.

RESULTS

The representations of the relationship: The theacher’s point of viewAs regards the links among the dimensions of the TRI, a first result con-

cerns the existence of a positive, significant association between “Self Per-spective Taking Dimension” and “Perspective Taking Dimension” (χ2= 7.04,


71

d.f. 1, p < .01). A low or high level in each dimension goes with a low or high level in each of the other dimensions, respectively.

AREAS TRI DIMENSIONS IRBI DIMENSIONS

ContentCompliance: teacher response refers to the child’s compliance with classroom (or other) rules.

Compliance: child response refers to the his/her compliance with classroom (or other) rules.

Achievement: the teacher’s references to the child’s performance of skills and behaviours or more general reference to developmental progresses.

Achievement: the child’s references to the his/her performance of skills and behaviours or more general reference to developmental progresses.

Secure Base: teacher acknowledges her/his role as a secure base for the child, allowing the child to actively explore and learn while know-ing that teacher is there to provide comfort and encouragement.

Secure Base: child acknowledges teacher’s role as a secure base, allowing him/her to actively explore and learn while knowing that teacher is there to provide comfort and encouragement.

Process

Perspective-Taking: teacher response indi-cates that s/he views the child as an individual with independent mental states, thoughts and feelings.

Perspective-Taking: child response indicates that s/he views the teacher as an individual with independent mental states, thoughts and feelings.

Self Perspective-Taking: teacher response in-dicates that s/he views him/herself as an indi-vidual with independent mental states, thoughts and feelings.

Self Perspective-Taking: child response in-dicates that s/he views him/herself as an indi-vidual with independent mental states, thoughts and feelings.

Neutralize Negative: teacher’s attempt to dis-tance him/herself from the negative affective component of the question.

Neutralize Negative: child’s attempt to dis-tance him/herself from the negative affective component of the question.

Agency: teacher’s feelings of confidence and effectiveness within the classroom.

Agency: child acknowledges teacher’s role of confidence and effectiveness within the class-room.

AffectiveTone

Positive: the teacher expresses internal experi-ences of positive affect.

Positive: the child expresses internal experi-ences of positive affect.

Negative: the teacher expresses internal expe-riences of negative affect.

Negative: the child expresses internal experi-ences of negative affect.

Domain Academic: any references to academic activi-ties, skills, achievement, etc. (reading, writing, math, etc.).

Academic: any references to academic activi-ties, skills, achievement, etc. (reading, writing, math, etc.).

Homelife: mentions home life (parents, sib-lings, baby-sitter, etc.).

Homelife: mentions home life (parents, sib-lings, baby-sitter, etc.).

TABLE 1. TRI and IRBI areas and dimensions

These data suggest that when a teacher, speaking about the relationship with his/her pupil, is able to deeply explicit his/her own perspective (that is the proclivity to speak about his/her own emotions, beliefs, mental states) he/she is also able to assume the pupil perspective. On the contrary, a school caregiver who has difficulty in accessing his/herself mental world is also not able to read the pupil’s mind. Moreover, a high level in “Self Perspective Tak-


72

ing Dimension” on TRI is linked both to a high proclivity to recognize his/herself as “Secure Base” for the child (χ2= 8.24, d.f. 1, p < .01) and to a high expression of positive affects (“Positive Affective Tone Dimension”) which emerge in the relationship with the pupil (χ2= 6.21, d.f. 1, p < .05).

Furthermore, the “Perspective Taking Dimension” is significantly linked with the expression of the “Negative Affective Tone Dimension” (χ2= 7.11, d.f. 1, p < .01): a high ability to assume the mental perspective of the child is connected with a rich expression of negative affects.

Finally, the “Neutralize Negative Dimension” is inversely connected both with the “Self Perspective Taking Dimension” (χ2= 10.49, d.f. 1, p < .01) and the “Perspective Taking Dimension” (χ2= 7.11, d.f. 1, p < .01). These data suggest that when the teacher is able to explore one’s own and other people’s mental world, he/she doesn’t elude the negative components which normally emerge in the relationship with the child.

The representations of the relationship: the child’s point of viewAs regards the links between the IRBI dimensions, a high level in the “Self

Perspective Taking Dimension” corresponds to a high level in the “Perspec-tive Taking Dimension”; and the opposite is also true (χ2= 12.43, d.f. 1, p < .001).

Thus, a child who is able to assume his/her own perspective is also able to read and verbalize other people’s mental states, just as difficulty in accessing one’s own inner world precludes the possibility to understand other people’s mental world.

Furthermore, the “Perspective Taking Dimension” is connected both with the “Positive Affective Tone” (χ2= 7.76, d.f. 2, p < .05) and with the “Agency Dimension” (χ2= 5.77, d.f. 1, p < .05): the children which profoundly assume the teacher’s perspective, in terms of mental states, also express deeply posi-tive affect and recognize the caregiver’s agency in his/her role of teacher.

Furthermore, a high level in the “Self Perspective Taking Dimension” and in the “Perspective Taking Dimension” is connected with: a) high level in the “Secure Base Dimension” (“Self Perspective Taking Dimension”: χ2= 6.20, d.f. 1, p < .05; “Perspective Taking Dimension”: χ2= 5.11, d.f. 1, p < .05); b) high level in the “Negative Affective Tone dimension” (“Self Per-spective Taking Dimension”: χ2= 11.09, d.f. 1, p < .001; “Perspective Taking


73

Dimension”: χ2= 5.6, d.f. 1, p < .05); c) low level in the “Neutralize Negative dimension” (“Self Perspective Taking Dimension”: χ2= 3.54, d.f. 1, p < .05; “Perspective Taking Dimension”: χ2= 8.58, d.f. 1, p < .01).

So when a child is a “good reader” of his/her own inner world (in terms of mental states), he/she also acknowledges the teacher’s role as a secure base, recognizing that the caregiver is available to provide comfort and en-couragement; his/her verbalizations are characterized by a high level of ex-pression of negative components of the relationship , which don’t need to be eluded.

Finally, the “Neutralize Negative Dimension” is significant and inversely correlated with the Negative Affective Tone Dimension” (χ2= 7.60, d.f. 1, p < .05) and with “Negative Compliance” ”(χ2= 4.86, d.f. 1, p < .05).

In particular, when a child eludes the negative component of the relation-ship, he/she is not able to assume the self and other person’s perspective, to express the negative affects which emerge in the relationship, or to verbalize the absence of compliance with classroom rules.

Attachment to teacher and child’s representationA one way Anova was performed assuming the quality of attachment to

teachers (SAT- School) as an independent variable and the scores on TRI dimensions as a dependent variable, showing that the quality of attachment to the teacher can influence the mean scores obtained by children in some IRBI dimensions.

The Newman Keuls test reveals that the insecure group obtained lower mean scores in the “Self Perspective Taking Dimension” ”(F(2 32)= 4.26, p < .05) and in the “Perspective Taking Dimension” (F(2 32)= 3.24, p < .05) than ambiguous and secure groups.

The results also show that the secure child is more able than the insecure child to recognize the teacher as a secure base (“Secure Base Dimension”: F(2

32)= 3.95, p < .05) and his/her role of confidence and effectiveness within the classroom (“Agency Dimension”: F(2 32)= 3.42, p < .05) (see table 2).


74

IRBI DIMENSIONS

INSECURE GROUP

(1)

AMBIGUOS GROUP

(2)

SECURE GROUP

(3)

F(2 32) POSTHOC(Newman-

Keuls Test’s)

SELF PERSPECTIVE

- TAKING4.22 4.83 6.64 4.26* 1 2 3

PERSPECTIVE-TAKING 4.31 5.85 6.87 3.24* 1 2 3

SECURE BASE 0.33 0.75 1.57 3.95* 1 2 23

AGENCY 1.78 3.75 4.50 3.42* 1 2 23

(* p< .05)

TABLE 2. SAT School attachment groups and mean scores in some IRBI dimensions. Anova and Post-Hoc

Attachment to parents and teachers and teacher’s representationsConsidering teachers’ verbalizations (TRI) and children’s attachment to

parents and teachers, the teachers reach higher levels in the “Negative Af-fective Tone Dimension” when they speak about secure groups rather than when they speak about insecure and ambiguous groups, as emerged from SAT-Family (χ2= 10.68, d.f. 4, p < .05) and SAT-School (χ2= 10.30, d.f. 42, p < .05).

In other words, the teachers are able to express their negative affect when the pupils they are talking about have a secure attachment both in the family and at school.

Child’s and teacher’s representations: the linksTable 3 presents the associations between the TRI and IRBI dimensions.

When a teacher is able to take on his/her own perspective, the pupil is able to


75

assume the teacher’s perspective. In other words, a teacher who is available to explore his/her own mental states allows his pupil to explore other people’s minds, assuming their perspective2.

Furthermore, there emerges a significant link in the “Neutralize Negative Dimension” between TRI and IRBI. When a teacher neutralizes the negative affects arising in the teacher-pupil relationship (TRI), the child also eludes the negative components of the relationship (IRBI).

The opposite happens when the teacher doesn’t neutralize the negative affects. So, it seems that when the teacher neutralizes the negative compo-nents, by eluding them, the child cannot explore the negative emotions and consequently eludes them.

TRIIRBI

PERSPECTIVE-TAKING

LOW HIGH TOTAL

SELF PERSPECTIVE -TAKING

LOW 15 (78.9%) 4 (21.1%) 19 (100%)

HIGH 7 (43.8%) 9 (56.3%) 16 (100%)

TOTAL 22 (62.9%) 13 (37.1%) 35 (100%)

χ2= 4.61, d.f. 1, p< .05

NEUTRALIZE NEGATIVE

LOW HIGH TOTAL

NEUTRALIZE NEGATIVE

LOW 13 (72.2%) 5 (27.8%) 18 (100%)

HIGH 4 (23.5%) 13 (76.5%) 17 (100%)

TOTAL 17 (48.6%) 18 (51.4%) 35 (100%)

χ2= 8.29, d.f. 1, p< .01

TABLE 3. Significant links between TRI and IRBI dimensions

2 This doesn’t mean interpreting our correlational data in a causal way, but rather believing that within a tutorial relationship the more competent partner directs the less competent one towards higher level abilities.


76

DISCUSSION

The results suggest the relevance of a consideration of the relational con-text in order to study the learning-teaching process: the relationship that a child builds with his/her school caregiver models his/her development (Pi-anta, 1999). In particular, through this study, we intended to underline that the teacher and pupil are not involved in a mere work of transmission and reception of knowledge, but they are actively involved in the construction of a relationship, developing through different –representational, affective, educational- dimensions. The aim of the research was to explore some of these dimensions.

As regards the teacher’s representations of the relationship with the pupil, as measured by TRI, the results suggest some interesting aspects. On the one hand, when the teacher is able to explore his/her inner world, he/she is also able to explore the child’s inner world and to express the positive affects he experiences towards him/her, and finally to recognize him/herself in his/her role of caregiver, that is the person that provides the child with a secure base. Moreover, the proclivity to conceive the other (in this case the pupil) as a person with mental states enables the expression (and not the elusion) of negative emotions too, which normally arise in the relationship.

These data seem to confirm the role of the caregiver’s mentalizing abil-ity, in terms of Fonagy’s (Fonagy & Target, 2001) and Meins’s (Meins, 1997; Meins at al., 2003) description.

As regards the child’s representations of the relationship with the teacher, as measured by IRBI, the analysis shows a link between the mentalizing abil-ity and the security of attachment. We are referring to security as the child’s recognition of the school caregiver as a secure base, who is available in every difficult situation. The children who have a good ability to assume perspec-tives, can also recognize and describe the teacher as an affective figure, who is able to confer security. Moreover, the ability to explore and to express his/her own and other people’s mental states makes the expression of negative af-fects possible (high level of “Negative Affective Tone Dimension”), without the need for elusion (Low level of “Neutralize Negative Dimension”).

As for the link between SAT-School and IRBI, the results support those theoretical positions (Fonagy & Target, 2001, Meins, 1997; Meins et al., 2003) that suggest there is a strong connection between the quality of attachment


77

and the child’s mentalizing ability. The secure children seem to be charac-terized by a high accessibility to their own and other people’s inner world. This evidence allows us to conceive the mentalizing ability not merely as the comprehension of false belief, as so frequently emerges from literature (Fonagy, Redfern & Charman, 1997; Meins, 1997). The present work points out a more general aspect of the child’s mentalistic ability, that is the ability to considerer the other (the teacher) as an individual with an independent mind, with mental states, feelings, emotions, that are not equal, but similar to his/her own.

Another interesting result concerns the link between the quality of attach-ment and the “Secure Base Dimension” and the “Agency Dimension” of the IRBI: the security of attachment allows the child to recognize the teacher both in his/her institutional role (the teacher who works in the classroom, establishing the rules and indications) and in his/her affective role (the secure base to which the child comes back in difficult situations and from which the child goes away to explore the world).

Analysing the links between TRI and SAT-School, it emerges that the teacher is also able to express a high level of negative affects when he/she speaks about the relationship with a secure child, both in the family and at school. These data can be interpreted in relation to the teacher-child rela-tionship by applying some concepts that Fonagy and Target (2001) adopt to understand the mother-child relationship. There is space for the negative component of the relationship only in the secure attachment context. The caregiver, involved in a secure relationship with the child, is able to contain and also to express him/herself and the child’s negative affects. In fact, in this case the caregiver is characterized by a good level of self reliance: he/she is trusting (or confident) in his/her own ability not only to contain these affects, but also to return them to the child, in a sensible way.

A global finding of the research concerns the presence of a dialogue be-tween the mind of children and teacher, as emerges from the TRI and IRBI dimensions. When the caregiver eludes the negative affects of the relation-ship, the child appears to be a “poor reader” of his/her own and other peo-ple’s minds. For this reason, the development of mentalization can be very complex and difficult. In other words, the adult’s inability to find in his/her own mind the negative affects experienced in the relationship and to express them (high level of “Neutralize Negative Dimension”- TRI) is accompanied


78

by the child’s refusal to explore his/her emotional negative experiences. This brings the child to neutralize the emotive-negative component of the rela-tionship (high level of “Neutralize Negative Dimension”- IRBI), not fully developing the mentalizing ability.

The results show that the teacher-child relationship represents a “meeting between minds”(Aron, 1996; Liverta-Sempio, 2002; Olson & Bruner, 1996), an intersubjective process where the recognition and the understanding of the partner’s emotions and thoughts could become a rich opportunity for learning and a protective factor for the child as well as an enrichment for both partners of the relationship.

REFERENCES

Aron, L. (1996). A meeting of minds, mutuality in psychoanalysis. Hillsdale, NJ: The Analitic Press.

Astington, J. (1996). What is theoretical about the child’s theory of mind? A Vygotskian view of its development. In P. Carruthers & P. Smith (Eds.) Theories of theory of mind (pp. 184-199). Cambridge, England: Cambridge University Press. In LIVERTA SEMPIO, O., MARCHETTI, A. (a cura di) Teoria della mente e relazioni affettive (pp. 13-33). UTET, Torino 2001.

Bion, W.R. (1962). Learning from experience. London: Heinemann.Bruner, J. (1986). Actual minds, possible words. Cambridge: Harvard University

Press.Bruner, J. (1990). Acts of meaning. Cambridge: Harvard University Press.Bruner, J. (1996). The culture of education. Cambridge: Harvard University

Press.Cutting, A., & Dunn, J. (1999). Theory of mind, emotion understanding,

language and family background: individual differences and interrelations. Child Development, 70, 853-865.

Dunn, J., Brown, J., Slomkowski, C., Tesla, C., & Youngblade, L. (1991). Young children’s understanding of other people’s feelings and beliefs: individual differences and their antecedents. Child Development, 62, 1352-1366.

Klagsbrun, M., & Bowlby, J. (1976). “Responses to separation from parents: a clinical test for young children.” British Journal of Projective Psychology and Personality study, 21, 7-26.


79

Fonagy, P., Redfern, S., & Charman, T. (1997). The relationship between belief-desire reasoning and a projective measure of attachment security (SAT). British Journal of Developmental Psychology, 15, 51-61. In: O. Liverta-Sempio & A. Marchetti (Eds.) (2001a) Teoria della mente e relazioni affettive (pp.153-176). UTET, Torino.

Fonagy, P., & Target, M. (2001). Attaccamento e funzione riflessiva. Milano: Raf-faello Cortina.

Lecciso, F., Liverta-Sempio, O., & Marchetti, A. (2002). La relazione educativa: la sua rappresentazione in insegnanti e allievi. Pre-atti del 4° Congresso Nazio-nale SIPEF “Gli psicologi nei contesti educativi: ricerca e formazione”, 5-7 Dicembre 2002, Urbino, 131-132.

Liverta-Sempio, O. (2002). Il rapporto educativo nell’ottica della psicologia socio-culturale. La Famiglia, 212, 16-28.

Liverta-Sempio, O., & Marchetti, A. (a cura di) (1995). Il pensiero dell’altro. Con-testo, conoscenza e teorie della mente. Milano: Raffaello Cortina.

Liverta-Sempio, O., & Marchetti, A. (a cura di) (2001a). Teoria della mente e relazioni affettive. Torino: UTET.

Liverta-Sempio, O., & Marchetti, A. (2001b) Prefazione all’edizione italiana. In R.C. PIANTA La relazione bambino-insegnante: aspetti evolutivi e clinici (pp. IX-XIV). Milano: Raffaello Cortina.

Liverta-Sempio, O., Marchetti, A., & Lecciso, F. (2001). Il SAT Famiglia e il SAT Scuola: strumenti di misura dell’ansietà di separazione da genitori e insegnanti. Milano: ISU, Università Cattolica del Sacro Cuore.

Liverta-Sempio, O., Marchetti, A., & Lecciso, F. (2002). L’Intervista sulla Rela-zione del Bambino con l’Insegnante (IRBI). Università Cattolica del Sacro Cuo-re, Milano – Brescia – Piacenza.

Liverta-Sempio, O., Marchetti, A., & Lecciso, F. (in press). Teoria della mente tra normalità e patologia. Milano: Raffaello Cortina.

Main, M., Kaplan, N,. & Cassidy, J. (1985). Security in infancy, childhood and adulthood: a move to the level of representation. In I. Bretherton, & E. Waters (Ed.), Growing points of attachment theory and research. Mono-graphs of the Society for Research in Child Development Vol. 50, 1-2, Serial No 209, 66-104.

Marchetti, A., & Massaro, D. (2002). Capire la mente. La psicologia ingenua del bambino. Roma: Carocci Editore.

Meins, E. (1997). Sicurezza e sviluppo sociale della conoscenza. Milano: Raffaello


80

Cortina, 1999. Meins, E., & Fernyough, C. (1999). Linguistic acquisitional style and mentalis-

ing development: the role of maternal mind-mindedness. Cognitive Develop-ment, 14, 363-380.

Meins, E., Fernyough, C., Fradley, M., & Tuckey, M., (2001). Rethinking ma-ternal sensitivity: mothers’ comments on infants’ mental processes predict security of attachment at 12 months. Journal of Child Psychology and Psychia-try, 42, 637-648.

Meins, E., Fernyough, C., Wainwright, R., Clark-Carter, D., Das Gupta, M., Fradley, M., & Tuckey, M. (2003). Pathway to understanding mind: con-struct validity and predictive validity of maternal mind-mindedness. Child Development, Vol. 74, 4, 1194-1211.

Olson, D.R., & Bruner, J.F. (1996). Folk psychology and folk pedagogy. In D.R. Olson, & N. Torrance (Eds.), The handbook of education and Human development. New modes of learning, teaching and schooling (pp.9-27). Cambridge: Blackwell.

Perner, J., Ruffman, T., & Leekam, S.R. (1994). Theory of mind is contagious: you catch it from your sibs. Child Development, 65, 1228-1238.

Pianta, R.C. (1999). Enhancing relationship between children and teachers. Ameri-can Psychological Association. Tr. It. O. Liverta-Sempio, & A. Marchetti (Eds.) (2001) La relazione bambino-insegnante: aspetti evolutivi e clinici. Milano: Raffaello Cortina.

Pianta, R.C., Bunosky, L., Fitz, M., Hamre, B., Kraft-Sayre, M., & Steimberg, D. (1999). Teacher Relationship Interview Coding Manual. University of Vir-ginia.

Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a theory of mind? Behavioral and Brain Sciences, 1, 515-526.

Repacholi, B., & Slaughter, V. (Eds.) (2003). Individual differences in theory of mind. Implications for typical and atypical development. New York and Hove: Psy-chology Press.

Slough, N.M., Goyette, M., & Greenberg, M.T. (1988). Scoring indices for the Seattle version of the separation anxiety test. University of Washington, Washington.

Vygotsky, L.S. (1978) Mind in society: the development of higher psychologi-cal processes. Cambridge: Harvard University Press.

Van Ijzendoorn, M.H., Sagi, A., & Lambermon, M.W.E. (1992) Il paradosso


81

del caretaker multiplo. Tr. it. in L. Carli (Ed.), (1999). Dalla diade alla famiglia (pp. 217- 238). Milano: Raffaello Cortina.

Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: representation and constraining function of wrong beliefs in young children’s understanding of deception. Cognition, 13, 103-128.

Winnicott, D. (1965). The maturational processes and the facilitative environment. London: Hogart.

Wood, D., Bruner, J., & Ross, G. (1976). The role of tutoring in problem-solving. Journal of Child Psychology and Psychiatry, 17 (2), 89-100.


83

EFFECTS OF EDUCATIONAL CULTURE AND TEACHING EXPERIENCE ON

TEACHERS’ BELIEFS

Ottavia Albanese*, Pierre-André Doudin**, Caterina Fiorilli***, Roberta Garbo*

INTRODUCTION

Implicit conceptions of intelligence are a very meaningful investigation and intervention field from the point of view of education and teacher train-ing. By using this term we intend to refer to the beliefs one has about the nature of intelligence and the factors which influence its development.

When we talk about implicit theories about intelligence we refer to be-liefs, conceptions, ideas one has in relation to issues like: what it takes to be intelligent and to show it, the way we interpret the efforts required to solve a given cognitive task, the importance we ascribe to errors, the way we react to failures we experience in tests and the opportunities we offer ourselves to become more intelligent (Dweck & Legget, 1988; Cherubini, 1988; Charnay & Mante, 1995; Doudin & Martin 1999a, 1999b; Dweck, 2000). We may be or may be not aware of such conceptions or beliefs which refer to our own intelligence or to other people’s. The distinction between implicit and explicit theories about intelligence refers to their very nature: the first are present in people’s minds, whether consciously or unconsciously, and belong to the cultural heritage of a given cultural group, the second have been constructed over time by the scientific community. Their effects can be assessed through the tools and tests developed to measure the value of intelligence in each sub-ject. The latter are explicit and shared, even though they are not necessarily the only ones we may encounter at a given time in history or within a given society (Sternberg, 1988; Farìa & Fontaine, 1994).

* University of Milano-Bicocca, Milano, Italy.** University of Genève, Switzerland.*** LUMSA University, Rome, Italy.

Ottavia Albanese et al.

84

The study of this topic has met with growing interest, especially for the impact these conceptions of intelligence have on a person’s actual behaviour. Meaningful contributions under this respect have come from the research team led by Dweck (1988, 2000) and their work on conceptions of intelli-gence and their influence on many aspects of personality: in particular, such an influence seems to be related to the orientation the conception of a given individual seems to take. Two prevailing orientations have been identified, one assuming intelligence is something defined from birth (entity theory), and one supporting the idea that abilities depend from progressive learning which makes it possible to increase intelligence (incremental theory).

With time, each individual develops ideas and beliefs about his/her own cognitive functioning, that is, the way he/she thinks about cognitive tasks, the meaning he/she should ascribe to achievements, and so on; such concep-tions pervade many aspects of a person’s cognitive life, especially learning objectives and motivation to learn. These conceptions, as we will see, change over time, as they are particularly responsive to the cultural context as well as to personal experience. Among the earlier studies on teachers’ conceptions of their pupils’ intelligence, those conducted by Mugny and Carugati (1985) highlighted that teachers tend to show different positions which could be represented as a continuum, with an innatist view at one hand and a con-structivist view at the other. According to the innatist view, also defined as the ideology of the gift (Bourdieu & Passeron, 1970) or of predestination (Chauvet & Gentil, 1993), the mind is little modifiable, intelligence is a fixed entity and achievements are due to a stable intellectual level. When entering school, pupils have their cultural, social and/or biological background which will determine their propensity to learn. In this sense, their performance will be basically interpreted by innatist teachers as the result of these initial char-acteristics which mark their pupils’ school experience. Teaching and learning will therefore play a marginal role, and any academic difficulty will first of all be interpreted on the basis of the pupil’s intrinsic characteristics with respect to family, social or innate aspects, leaving out from the analysis of possible sources of difficulty all teacher- and school-related variables.

On the contrary, the constructivist view considers intelligence as the result of multiple factors, subject to change and improvements, among which learn-ing plays a major role. Those who think about intelligence in these terms, even though implicitly, ascribe to the child, the teacher and to contextual variables

Effects of educational culture and teaching experience on teachers’ beliefs

85

a key role in the development of intelligence. Under this respect, the teacher-pupil relation plays an active role in the co-construction of knowledge (Alba-nese, Doudin and Martin, 1999; Albanese and Fiorilli, 2000); furthermore, in case of school failure, the teacher will be more inclined to consider the event as linked to the failure of teaching-learning processes (Seligman, 1991; Bres-soux, 1994). In the field of educational research, growing attention has been devoted to the study of the teacher’s educational practice as a function of his/her conception of intelligence, with special emphasis on the effects it can have on the child’s performance and feeling of self-efficacy. Based on the re-sults, one is led to assume that teachers with different belief systems manage teaching-learning processes differently (Doyle, 1986; Good, 1982; Thomp-son, 1992; Seligman, 1991; Bressoux, 1994; Czaja, 1999; Bandura, 2000).

The innatist teacher tends to consider the educational practice as some-thing with little impact and meaning for development. Such a conception is expressed in teacher-pupil interactions where closed questions are prevailing, little room is left to the pupil’s free search, there is a tendency to identify the sources of school failure outside the educational process as such, and to con-sider failure as an effective educational measure. In contrast, the constructiv-ist teacher, who attaches greater importance to the progressive function of learning, chooses to develop interactions where pupils are more intensely stimulated towards the construction of knowledge, for instance through the use of open questions and the practice of group discussion, and difficulties are interpreted as steps towards the acquisition of a given competence (Mar-tin, 1991; Bressoux, 1994; Czaja, 1999).

In particular, teachers’ conceptions depart sharply when it comes to inter-preting the errors made by pupils during learning activities. Allal and Schu-bauer-Leoni (1992-1993) have analysed the meaning ascribed to errors in the two conceptions through interviews, observing that the innatist teacher views errors as signs of the pupil’s intrinsic difficulty while the constructivist teach-er interprets them as difficulties in the teaching-learning process.

This difference in interpretation stems from a more general attitude, that is, to consider school failure as a result of the pupil’s initial condition in an innatist perspective, or else to view it in a constructivist framework as gener-ated by the combination of several variables, some of which linked to the child’s inborn and family characteristics, some others connected to the school context. But what are the factors that lead teachers to adopt innatist rather


86

than constructivist conceptions?As shown by the studies by Mugny and Carugati (1985) on the represen-

tations of intelligence in different social groups, teachers and parents, social position and educational role meaningfully influence expectations related to pupils’/children’s learning, the sense of responsibility vs. failure and success and, in general, the beliefs about the possibility to increase intelligence.

The effects of social roles and responsibilities related to pupils’ learning were reported also in a previous study of ours, where we observed that the number of years in service plays a role in changing teachers’ conceptions (Al-banese & Fiorilli, 1999, 2000). In particular, teachers with over five years of service tend to take an innatist perspective, whereas teachers with little or no experience consider their pupils’ intelligence in constructivist terms. This shift of beliefs towards more conservative assumptions, after some years of teach-ing experience, may be explained by the social and sharing process teachers go through: personal beliefs would connect with those more specifically linked to the educational context owing to the fact that teachers gradually take on an identity associated with their role. The ideas which are typical of constructiv-ist teachers at the beginning of their careers tend to decline when they are faced with pupils’ failures leading them to take on a defensive attitude and to become entrenched in an innatist conception (Deschamps, 1992; Perrenoud, 1993; Gosling, 1992; Ross, 1996). As seen above, teachers with an innatist view of intelligence, explain school failure with variables connected to the pupil’s abilities or to his/her family and, at any rate, to factors that are beyond their control. To sum up, a connection can be found between a certain type of con-ception and educational practices. In particular a teacher who is prevailingly innatist in his/her beliefs is more likely to go for a different educational prac-tice compared to a teacher with a rather constructivist view. In addition, teach-ing experience has a negative influence on teachers’ beliefs as it leads them to take on, with time, an increasingly innatist perspective on intelligence.

THE RESEARCH

We hypothesised that the trend of experienced teachers to adopt a rather innatist perspective on their pupils intelligence would be recorded even in different educational cultures. In order to test our hypothesis we conducted


87

a research project with Italian and Swiss teachers, with and without teaching experience, starting from the basic assumption that this two groups belong-ing to different educational cultures.

Participants, method and procedureThe participants are 373 teachers, 161 from Italy (Siracusa) and 212 form

Switzerland (Fribourg). In each group, teachers were distinguished in experi-enced, on average teachers who had over 5 years of teaching experience, and non-experienced. While the experienced group taught different subject-mat-ters in Elementary Schools, those without professional experience were en-gaged during their university training or professional courses for pre-service teachers. Most teachers were female (see Table 1).

Italy Switzerland Tot

Experienced Teachers 101 100

Non-Experienced Teachers 60 112

Tot Participants 161 212 373

TABLE 1. Participants broken down by country and experience

A questionnaire on teachers’ beliefs, including 75 items regarding ideas about the development of intelligence, pupils’ failure, social relations with peers and adults and educational practice formulated by Pfulg (2000), was translated into Italian and administered to Italian teachers by Albanese and Fiorilli (2001). The questionnaire included items related to a constructivist conception and items related to an innatist conception: for each item, teach-ers express their opinions on a 6-point Likert scale that goes from strongly disagree (1) to strongly agree (6). For each subjects we obtained a final score by giving a negative sign to innatist items and a positive sign to constructivist items, then calculating the algebraic sum all items. The study involved two stages: the first aimed at validating the Italian version of the selected research tool, the second was meant to compare the Italian and Swiss teachers’ an-swers to the validated questionnaire.


88

Results1) Validation of questionnaireWe conducted a Rasch Model Analysis so as to check each of the original

75 items for fit with Rasch Model. Considering unacceptable chi-square dif-ferences over + 1.64 with p<.05 between expected and observed responses for each item, we removed items with a bad fit to the Rasch Model, and we thus obtained an abridged version of the questionnaire composed 29 items. After running the test several times we obtained a good global reliability (χ2= 62.597, d.f. = 72, and p>.778) that means a good fit between the Rasch Mod-el and the questionnaire (see Table 2).

Items No. Chi-square p1 0.264 .8732 1.556 .4453 1.737 .4044 3.003 .2025 2.357 .2896 .457 .7907 2.357 .2898 3.018 .2009 2.061 .34010 1.829 .38511 .347 .83612 .328 .75113 1.137 .55514 3.003 .20215 .932 .27916 3.670 .52117 .995 .59818 3.458 .29119 3.050 .19720 .193 1.05221 .274 .79222 3.420 .25023 .284 .86424 1.223 .53025 6.580 .06526 1.499 .45827 .382 .82128 .106 .94729 .444 .796

X2=62,597, gdl=72, p>.778

TABLE 2. Selected items on the basis of Rasch Model Analysis


89

We transformed the original questionnaire in the Scale of Constructivist Conception of Intelligence Development (SCCID) (see Appendix 1).

2) Effects of teaching experience in two different educational culturesIn order to examine the effect of their own educational culture on teach-

ers’ responses we compared the scores obtained on the SCCID by Italian and Swiss teachers. The first one-way analysis between Italian and Swiss teach-ers revealed no significant differences (see Table 3), indicating that teach-ers’ conceptions about their pupils’ intelligence is not affected by educational culture.

We compared the two groups of experienced and non-experienced teach-ers within each culture. Two one-way ANOVAs revealed significant differ-ences between experienced and non- experienced Italian teachers (F(1.161) = 8.541, p < .001) and between experienced and non-experienced Swiss teach-ers (F(1.212) = 9.690, p<.001), (see Table 3).

In both comparisons, non-experienced Italian teachers (M = 2.326; SD = 0.72) and non-experienced Swiss teachers (M = 2.307; SD = 0.71) show a significant tendency to think the development of their pupils’ intelligence in more constructivist terms than their more experienced colleagues in Italy (M = 2.165; SD = 0.69) and in Switzerland (M = 2.134; SD = 0.83) (see Table 3).

Italian Teachers, Experienced Italian Teachers, Non-Experienced Variance Analysis

M D.S. M D.S. F df

2.165 .69 2.326 .72 8.541** 1,161

Swiss Teachers, Experienced Swiss Teachers, Non-Experienced Variance Analysis

M D.S. M. D.S. F df

2.134 .83 2.307 .71 9.690** 1,212

*p<.05; **p< .01; ***p<.001

TABLE 3. Teachers’ scores on SCCID. Averages and differences

No significant differences were found between Italian and Swiss groups, but only within the same nationality group (see Graph 1).


90

Key: T_NE_IT: Non-Experienced Italian TeachersT_E_IT: Experienced Italian Teachers T_NE_SW: Non-Experienced Swiss TeachersT_E_SW: Experienced Swiss Teachers

GRAPHIC 1. Average score on SCCID of Italian and Swiss Teachers (Experienced and Non Experienced)

DISCUSSION AND CLOSING REMARKS

The beliefs of two groups of teachers, Italian and Swiss, including two sub-groups of experienced and non-experienced teachers were analysed in this study. Aim of the study was to investigate to what extent the belief sys-tem of two groups of teachers was influenced by educational culture and/or by their belonging to a specific social group, namely that of teachers. The results of our study show a tendency to take on an innatist conception of intelligence alongside with teaching experience and independently of educa-tional culture. Besides the cultural variable, another factor and just as much powerful comes into play, that is the need to defend one’s own role and pro-fessional identity. In fact, non-experienced teachers in both cultures show constructivist conceptions of intelligence whereas teachers with at least five years of teaching experience seem to have a tendency towards a static view of intelligence.

Thus, a cultural relativism can be observed when identifying dimensions

Aver

age

scor

es

T_NE_IT T_E_IT T_NE_SW T_E_SW

Teachers group

3

2,5

2

1,5

1

Teachers’ answers on the SCCID


91

of intelligence (Sternberg et al., 1981) even though under certain circum-stances beliefs are constructed and transmitted which are implicitly shared by a group and connected to a social rather than cultural identity (Farìa & Fontaine, 1997, 1994; Schurmans, Dasen & Vouilloz, 1990).

The interpretation of these results leads us to consider the factors that play a role in the transformation of teachers’ conceptions and help bring to light the importance of in-service training for teachers. The point we would like to stress concerns the possibility to work on teachers’ beliefs and favour change processes exactly where and when they seem to have a negative influ-ence on teaching-learning processes (Battie, 1995; Calderhead, 1993; Doudin & Martin, 1999a, 1990b; Ferguson, 1989).

Training programmes are bound to fail when they ignore the existence of a belief system, as well as of pre-existing knowledge in teachers. More than this, in order to be effective, they must choose training tools which may elicit beliefs, making them visible and therefore modifiable (Anderson et al., 1995).

Guskey (1986) found that programmes meant to favour changes in teach-ers’ beliefs and attitudes fail if they are based on providing information, while they tend to have comforting results when teachers are invited to use a new procedure and to evaluate its positive impact on the performance of their pupils. This change does not occur when teachers do not use the new sug-gested strategies actively or when, even though they use them, they cannot observe improvements in their pupils. Starting from this finding, the author contends that the change in beliefs follows, rather than preceding the change in behaviours.

A training programme may provide a source of information and new knowledge, may help participants become aware of the fact that there is a different knowledge from their own beliefs and still leave such beliefs un-changed. An intervention programme aimed at “correcting” teachers’ im-plicit theories towards an incremental conception of intelligence showed an immediate positive effect on the sense of self-efficacy of participants which however disappeared in the post-test (Nettle, 1998). The training tool em-ployed relied heavily on verbal persuasion which is a weak source for chang-ing beliefs (Bandura, 1981, 1996).

Thus, different studies on the effects of training programmes meant to favour change in teachers’ implicit conceptions have shown that pre-existing


92

beliefs and new knowledge may continue to coexist even when they are not consistent with each other (Joram & Gabriele, 1998; Nespor, 1987). Training programmes often focus on the acquisition of explicit rational techniques, therefore leaving little room for analysing teachers’ personal and implicit trends. A training programme which could take these aspects into due ac-count would therefore focus on making these conceptions explicit. The ob-jective, in this case, would be one of using teachers’ personal beliefs about pupils’ intelligence as a source of understanding about themselves and their own thinking. In order for new pieces of information to be used to change and accommodate previous beliefs, a person should: a) recognise that the new piece of information can somehow be brought back to existing beliefs; b) be aware that assimilation efforts are unsatisfactory and that the new ele-ment challenges existing data; c) wish to reduce the inconsistency between these elements (Posner et al., 1982).

Changing beliefs thus calls for a metacognitive process that, through re-flection, may highlight the nature of beliefs and their difference with respect to knowledge (Jarom & Gabriele, 1998).

A possible teacher training format could involve two steps: 1) support teachers’ reflective thinking on their practice and its effects, thereby favour-ing their critical self-observation skills; 2) evaluate effectiveness of alternative strategies for management and transmission of knowledge, for instance hav-ing experienced and non-experienced teachers exchange their views about possible ways to solve specific problems (Martin, 1991; Martin & Doudin, 1998; 2000; Pallascio & Lafortune, 2000; Paquay, 2000).

In terms of methodology, in-service teacher training trough self-regula-tion, self-evaluation and exchange among peers seems to emerge as the way which will more likely facilitate an appreciation of the differences between knowledge and beliefs and the experimental practice of new teaching condi-tions (Albanese, Doudin & Martin, 2003).

REFERENCES

Albanese O. and Doudin P.-A. (1999), L’interazione in classe: quale intera-zione per un apprendimento efficace, in O. Albanese, P. Migliorini, G. Pietrocola (eds.), Apprendimento e nuove strategie educative, (pp. 22-36), Milano:


93

Unicopli.Albanese O. and Fiorilli C. (2000), Ruolo dell’esperienza e della cultura nelle

concezioni degli insegnanti, Atti del III Congresso Nazionale SIPEF, Profes-sione e ricerca in Psicologia dell’educazione e della formazione, (pp. 22-23), Trento: Ed. Erickson.

Albanese O. and Fiorilli C. (2001), Scuola e concezioni dello sviluppo del-l’intelligenza negli insegnanti, XV Congresso Nazionale AIP, (pp. 146-147), Palermo, 26-28 settembre, 2001.

Albanese O., Doudin P.-A. and Martin D. (eds.), (2003), Metacognizione ed edu-cazione, Milano: Franco Angeli.

Allal L. and Schubauer-Leoni M.-L. (1992-1993), Progression scolaire des élèves: le redoublement dans le contexte genevois, Théorie et Practique, 11-12, 41-50.

Anderson L.M., Blumenfeld P., Pintrich P.R., Clark C.M., Marx R.W. and Peterson P. (1995), Educational psychology for teachers: Reforming our courses, rethinking our roles, Educational Psychologist, 30, 143-157.

Bandura A. (1981), Self-referent thought: A developmental analysis of self-efficacy, in J. Flavell and J. Ross (eds.), Social cognitive development: Frontiers and possible futures, (pp. 200-239), Cambridge: Cambridge University Press.

Bandura A. (2000), Autoefficacia: teoria e applicazioni, Trento: Ed. Erickson.Bandura A. (eds.), (1996), Il senso di autoefficacia, Trento: Ed. Erickson.Battie M. (1995), New prospects for teacher education: narrative ways of

knowing teaching and teacher learning, Educational Research, 37, 1, 53-70.Bourdieu P. and Passeron J.C. (1970), La reproduction : éléments pour une théorie du

système d’enseignement, Paris: de Minuit.Bressoux P. (1994), Les recherches sur les effects-école et les effets-maïtres,

Revue Française de Pédagogie, 108, 91-137.Calderhead J. (1993), The contribution of research on teachers’ thinking to

the professional development of teachers, in C. Day, J. Calderhead and P. Denicolo (eds.), Research on teacher thinking: Understanding professional develop-ment, (pp. 291-310), New York: Macmillan.

Charnay R. and Mante M. (1995), Préparation à l’épreuve de mathématiques du concours de professeur des écoles, Tome I, Paris: Hartier.

Chauvet J. and Gentil R. (1993), Les représentations liées à l’expression “élève en difficulté”, Education et Formation, 36, 13-20.

Cherubini G. (1998), La ricerca sulle conoscenze e sulle credenze degli inse-


94

gnanti: quali le implicazioni per la loro formazione e lo sviluppo profes-sionale? Rassegna di Psicologia, 15(1), 51-62.

Crahay M. (1996), Peut-on lutter contre l’échec scolaire ? Bruxelles: De Boeck.Czaja C.F. (1999), «The effect of teacher style on interactive engagement»:

Comment, Early Childhood Research Quarterly, Vol 14(1), 71-74.Deschamps J.-C. (1992), L’échec scolaire : une perspective attributionnelle en

psychologie sociale, in B. Pierrehumbert (ed.), L’échec scolaire: échec de l’école, (pp. 49-79), Neuchâtel et Paris: Delachaux & Niestlé.

Doudin P.-A. and Martin D. (1999a), Conception du développement de l’in-telligence et formation des enseignants, Revue Française de Pédagogie, 126, 121-132.

Doudin P.-A. and Martin D. (1999b), Métacognition, échec scolaire et forma-tion des enseignants, Actualités psychologiques, 6, 15-33.

Doyle W. (1986), Classroom organization and management, in M.C. Wittrock (ed.), Handbook of Research on Teaching (III ed.), (pp. 392-431), New York: MacMillan,

Dweck C. (2000), Self-Theories: Their role in motivation, personality and development, London: Taylor & Francis.

Dweck C. and Leggett E.L. (1988), A social-cognitive approach to motivation and personality, Psychological Review, Vol. 95, 256-273.

Farìa L. and Fontaine A.M. (1994), Teacher’s social representations of intel-ligence: An exploratory study with Portuguese teachers, in A. Clemente, G. Musitu and G. Gutiérres (eds.), Intervenciòn Educativa e Desarrolo Humano, (pp. 29-38), Valencia: Set i Set Edic.

Farìa L. and Fontaine A.M. (1997), Adolescents’ personal conceptions of intelligence: The development of a new scale and some exploratory evi-dence, European Journal of Psychology of Education, XII, 1, 51-62.

Ferguson P. (1989), A reflective approach to methods practicum, Journal of Teachers Education, 40, 36-41.

Good T.L. (1982), How teachers’ expectations affects results, American Educa-tion, 18, 25-32.

Gosling P. (1992), Qui est responsable de l’échec scolaire ? Représentations sociales, attributions et rôle d’enseignant, Paris: PUF.

Guskey T.R. (1986), Staff development and the process of teacher change, Educational Researcher, 15(5), 5-12.

Joram E. and Gabriele A.J. (1998), Pre-service teachers’ prior beliefs: Trans-


95

forming obstacles into opportunities, Teaching and Teacher Education, Vol. 14(2), 175-191.

Martin D. (1991), (Méta)communiquer pour apprendre, c’est faire de l’oral à plein temps, in M. Wirthner D., Martin and Ph. Perrenoud (eds.), Parole étouffée, parole libérée : fondements et limites d’une pédagogie de l’oral, (pp. 205-219), Neuchâtel et Paris: Delachaux & Niesté.

Martin D. and Doudin P.-A. (1998), Métacognition et formation des ensei-gnants, in L. Lafortune, P. Mongeau and R. Pallascio (eds.), La métacognition et les compétences réflexives en éducation, (pp. 23-46), Montréal: Les Edition Lo-giques.

Martin D. and Doudin P.-A. (2000), De l’utilité de la pensée réflexive pour améliorer l’efficacité de l’école, in R. Pallascio and L. Lafortune (eds.), Pour une pensée réflexive en éducation, (pp.47-67), Laurier, Sainte-Foy, Québec Univerisity Press.

Mugny G. and Carugati F. (1985), L’intelligence au pluriel. Les représentations socia-les de l’intelligence et de son développement, Cousset, Suisse: Delval.

Nespor J. (1987), The role of beliefs in the practice of teaching, Journal of Curriculum Studies, 19(4), 317-328.

Nettle E.B. (1998), Stability and change in the beliefs of student teachers’ during practice teaching, Teaching & Teacher Education, Vol. 14(2), 193-204.

Pallascio R. and Lafortune, L. (2000), Pour une pensée réflexive en éducation, Qué-bec: Québec University Press.

Paquay L. (2000), Quand un stage vise la diffusion de dispositifs innovants et la professionnalisation des enseignants, in G. Carlie, J.-P. Renard and L. Pa-quay (eds.), La formation continue des enseignants. Enjeux, innovation et réflexivité, (pp. 155-197), Bruxelles: De Boeck.

Perrenoud Ph. (1993), Echec scolaire: de la suppression du redoublement à la différenciation de l’enseignement, un long chemin, Recherche en étudiants : théorie et pratique, 16/17, 3-23.

Pfulg L. (2000), Les représentations des enseignants et étudiants de l’Ecole Normal du canton de Fribourg à propos du développement de l’intelligence, de l’échec scolaire et des pratiques pédagogiques, Faculté des lettres de l’Université de Fribourg (Swit-zerland), (unpublished manuscript).

Posner G.J., Strike K.A., Hewson P.W. and Gertzog W.A. (1982), Accommo-dation of a scientific conception: Toward a theory of conceptual change,


96

Scientific Education, 66, 211-228.Ross J.A. (1996), Strategies for enhancing teachers’ beliefs in their effective-

ness: Research on a school improvement hypothesis, Teachers College Record, 6, 226-251.

Schurmans M.-N., Dasen P. R. and Vouilloz M.-F. (1990), Composantes des représentations sociales de l’intelligence: Kpouebo (Cote d’Ivoire) et Evo-lene (Switzerland), in N. Bleichrodt and P. J.D. Drenth (eds.), Contemporary issues in cross-cultural psychology, Berwyn, Pa: Swets & Zeitlinger.

Seligman M.E.P. (1991), Learned Optimism, New York: Knopft.


97

APPENDIX

Scale of Constructivist Conception of Intelligence Development (SCCID)

Strongly StronglyDisagree Agree

1 2 3 4 5 6

1 The growth of intelligence develops according to a program biologically determined from birth

2 When a student is enable to performer in class one of the reasons might be in the fact that teaching methods were not stimulating appropriately

3 Interactions among companions in the class constitute a pow-erful catalyst affecting students’ progress

4 Student experiences failure because his/her family does not give him/her the opportunity to do well

5 Methods aimed at favouring intellectual development have proven, for the most part, ineffective: differences in perform-ance are rooted in social and cultural variants

6 The student develops his intellectual capacities by means of his/her interaction with others and with environment

7 Intellectual differences among individuals are accentuated in the school environment

8 Teachers share in the responsibility regarding the lack student progress

9 Errors constitute a new starting point for student progress

10 The affective atmosphere in the family of the student is the root cause of student failure

11 The intellectual development occurs in successive stages; each one requiring the re-organisation of previously acquired abili-ties

12 In the absence of congenital deficits, the underlying cause of defects in the intellectual development can be traced to the family history

13 The school allows the students to develop their own intelli-gence

14 The negative or positive expectations of the teacher influence student performances

15 From the ages of 12-14 years there is not further development in intelligence

16 The student plays a major role in any of the scholastic difficul-ties in which he/she may eventually find him/herself


98

17 Intelligent students cannot come from families in which learn-ing is not valued

18 Student failure is first of all the result of failure on the part of the school

19 In order for a group-work activity to be effective it is necessary that the children have the same intellectual level

20 The more a student receives praise the more he/she feels val-ued and therefore makes progress

21 Intellectual potential is innate, it cannot be altered

22 Intelligence is re-educable and the role of the teacher is fun-damental

23 The quality of the teacher-student relationship is very influ-enced on the scholastic performance of the students

24 If children are given opportunities to work together they will develop their intelligence better than if they worked individu-ally

25 Scholastic maladjustment is generally linked to family prob-lems

26 The frequent repetition of exercises is indispensable for the development of intelligence

27 Differences in intelligence highlighted by the school learning experiences are due to the diversity of the family social envi-ronment

28 To further normal intellectual development, a student must have the assurance of being able to steer the course of his/her own future

29 If students of some particular social classes seem more intel-ligent it is because these classes define what intelligence is and impose definition on others

Items that measure contructivist conception of intelligence are: 2, 3, 6, 7, 8, 9, 11, 13, 14, 18, 20, 22, 23, 24, 28.Items that measure innatist conception of intelligence are: 1, 4, 5, 10, 12, 15, 16, 17, 19, 21, 25, 26, 27, 29


99

APPROACH TO LEARNING IN THE PRIMARY SCHOOL: MATHEMATICS COMPARED TO

LANGUAGE INSTRUCTION

Manuela Cantoia*

This study is aimed at probing the differences in learning conceptions between two different school matters: language and mathematics. Do the different experiences and the different structures of these subjects allow students to develop different conceptions of learning? 188 children of 8-10 years of age (Mathematics Group: 75; Language Group: 113) received six short stories each followed by multiple-choice questions. Conceptions of learning in the two domains can vary according to the style of the teacher, the relation between student and teacher or the metacogni-tive competence of the students. Mathematics representations are more affected by school performance, gender and class variable.

INTRODUCTION

Conceptions of learning are characterised by a set of features, which may help to describe the student’s approach as a whole. The ideas that students develop about learning since their first years of school depend on motivation, strategies and behaviours, attribution, sense of self-efficacy, and what we could refer to as the “sense-making” of their school experience (Säljö, 1979; Marton, 1988; Mezirow, 1990; Entwistle & Marton, 1994). Conceptions of learning represent a main issue in psychology of learning and in developmen-tal psychology: their early appearance and their influence on the approach to studying and schooling has been assessed by many authors (Pramling, 1988; Van Rossum e Schenk, 1984; Campbell, 2001; Gibbs, 1995).

* Department of Psychology, Catholic University of the Sacred Heart. Email: [email protected]

Manuela Cantoia

100

During these last years, conceptions of learning have been probed in many ways, for instance by investigating the beliefs of the students, by matching stu-dents’ and teachers’ opinions, by assessing their developmental trend (Gow & Kember, 1993; Klatter et al., 2001). Researches on this field could help us to afford important issues as cognitive development, metacognition, school assessment and level of performance (Martin e Ramsdem, 1987; McLean, 2001).

One can maintain that conceptions of learning could be described as the core from which the whole school experience of each student takes its de-parture. However, it is somehow surprising that there is almost no literature about the comparisons among conceptions of learning in different school matters. Klatter et al. (2001) referred in a very general way to a significant relation between a developmental learning conception and the interest of primary school students regarding subjects like language and mathematic: as more developed the conceptions of learning are, more positive is the motiva-tion for school and learning. In Italy there is a long tradition of studies about emotional, motivational, strategic issues in single school domains (Cantoia et al. 1997; Lucangeli & Passolunghi, 1995; Mason & Scrivani, 2004; Poli & Zan, 1996; Pellerey, 1996), about general representations and epistemic be-liefs (Mason, 2001), and also about pedagogical practises and their influence on the performance of the class, but almost all the researches focused on only one matter at time.

When we talk about “learning” in the most general sense of the word, we can refer to a lot of literature, but there is the real risk to refer to too general stances: school experience may vary according to different elements as the context, the teachers, the source materials (What materials do students use? What kind of books, of technology? Do they affect students’ experi-ence? How?), pedagogical choices (learning by discovery, directive lessons, laboratory, etc.). To make this point clearer, we can quote a previous study1 on school experience during the primary school. Students were presented an imaginary situation describing the reflections that two children develop about their personal learning experience at school. Respondents were asked

1 Cantoia M., Antonietti A., La rappresentazione dell’apprendimento, unpublished work. The instrument used in this study is described in Antonietti (1998, pag. 32).

Approach to learning in the primary school

101

to choose among three different possible ways to figure out how the situation can develop: the first alternative is focused on the representation of strain, the second one on the sense of absurd and the third one on amusement. Children were also asked to complete the end they had chosen for the situ-ation by telling a personal school experience. The third alternative (amuse-ment) was selected by the 76.5% of the sample (102 students of 8-10 years of age) who coherently talked about good, positive experiences, personal satisfaction, friendships, school trips, stimulating lessons, etc. Therefore, it is possible to infer a positive attitude towards school and learning, but it is also possible to argue that probing the “school” as a whole could take to extreme generalizations or to underline just the relational or anecdotic aspects.

This study is aimed at prove two different and evocative domains – math-ematics and language (Italian) – to let the differences in the representations of a few basic elements emerge. Our hypothesis is that students’ ideas about learning vary when they approach and study different school matters. Thus we tried to get a specific picture of the hypothesized differences in the ap-proach to learning.

AN EMPIRICAL STUDY

Participants2

The sample in this study was constituted of 188 students of primary school (grade III, IV and V) of two different schools in the neighbours of Milan, which were comparable according to the social and geographic level. Twelve classes took part in the study (four classes for each grade) and they were divided in two groups (mathematics and language). Both genders were equally represented in the two groups, even if studying class-unit does not allow operating a complete balance of the sample according to gender. A be-tween group study was preferred because of the characteristics of the mate-rial: a double administration in one time would have taken to a learning effect because of the similar structure of the trials (the texts varied, but the multi-

2 I would like to thank Silvia Anghileri and Marta Di Candia who collect the data.

Manuela Cantoia

102

ple choice questions were similar) and two administrations in two different moment could have generated maturity effects, because children could have experienced significant situations during the school year. Table 1 illustrates the sample.

Mathematics Language

III IV V III IV V

M F M F M F M F M F M F

12 15 14 8 12 14 15 22 20 17 25 14

27 22 26 37 37 39

75 113

188

TABLE 1. Distribution of the sample

Materials and procedureThis study had a reflective purpose that has been stressed by using a meta-

cognitive instrument, which deepens the most significant characteristics in the relation with a school matter: skills, motivation, learning goals, strategies, attribution and learning behaviours3.

Each class was presented six short stories about a situation of school life during classes of mathematics/language. After having read a story, children had to answer individually to three multiple-choice questions whose alterna-tives (from 4 to 9 according to the story at hand) had been devised on the basis of literature on the subject. Children had to choose both the most and the least important alternative, to show how they would have act in the same situation, and to point out what elements could have been learned (innate vs. developmental perspective). The six stories, each focused on a different issue (skills, motivation, requests, behaviours, strategies, attribution), were arranged

3 The materials used for the Language sub-sample were published in Antonietti and Cantoia (2000); data of the Mathematics sub-sample have been discussed in Cantoia M., Matematica e metacognizione: alunni ed insegnanti della scuola elementare a confronto, IV Convegno Nazionale Sipef, Urbino, 5-7 December 2002.


103

into a booklet divided in three sections (see Table 2). Data about the first multiple-choice question will be discussed. The question about the most/least important element asked “What (issue at hand) of the characters do you think is the most important? And what is the least one?” Then children were presented a numbered list of sentences that were exactly those by which the single alternatives had been introduced in the text and they had to write the two numbers of their choices. In the first question children didn’t face any theoretical term, the alternatives were presented by referring to the context in which they had been previously explained. The single issues had been ex-plained to the children as follows: “One need to be able to…” (skills), “One studies (school subject) for…” (motivation), “What do one think he has to do?” (learning goals), “What do one do when he studies (school subject)?” (strategies), explications (attribution) and behaviours (learning behaviours).

Section of the booklet Story/Issue

1 Skills

2 MotivationGoals-teacher requests

3 Learning behavioursStrategies

Attribution

TABLE 2. Materials

The administration lasted one session (two sessions only for a few classes of the III grade) in the presence of a teacher-experimenter who worked in the school but was not involved in the teaching teams of the examined classes.

DATA ANALYSES AND RESULTS

In the present paper only responses given to the first kind of request (“What is the most/least important ...?) were analysed.

Chi square tests were computed by considering the distributions of the choices of the alternatives with respect to each story (each story correspond-

Manuela Cantoia

104

ed, as told previously, to a specific issue), both in response to the request to select the most important and the least important alternative. Table 3 shows that in all cases the distribution of the choices (by collapsing the two subject matters) revealed a significant concentration on one or more alternatives.

Factors N df Chi square

Important skill 186 6 119.892***

Unimportant skill 186 6 167.613***

Important motivation 188 4 110.032***

Unimportant motivation 188 4 152.213***

Important learning goal 187 6 137.358***

Unimportant learning goal 187 6 136.383***

Important learning behaviour 188 3 99.021***

Unimportant learning behaviour 188 3 64.492***

Important strategy 188 8 63.904***

Unimportant strategy 188 8 56.819***

Important attribution 188 8 258.553***

Unimportant attribution 188 8 95.213***

*** = p <.001

TABLE 3. Chi square test values for each question

Successively the above mentioned distributions were crosstabulated by considering the school matter (mathematics vs. language). The two sub-groups were analysed separately according to the grade (III vs. IV vs. V), gender (male vs. female) and school performance as assessed on the basis of a synthetic evaluation expressed by teachers (not sufficient, sufficient, good, very good, optimum). Significant data will be discussed by considering each of the six issues.

SkillsChildren were asked to show the most important/unimportant skill among

seven possible alternatives: attention, comprehension speed, memory, intel-


105

ligence, reflection, engagement, interest for the matter (crosstabulation be-tween alternatives chosen and matter: important X2(N = 186) = 45.628, df = 6, p < .001; unimportant X2(N = 186) = 26.281, df=6, p < .001) (Table 4).

IMPORTANT SKILL UNIMPORTANT SKILL

MG LG MG LG

Concentration 45.2 25.7 1.4 1.8

Memory 2.7 8 15.1 6.2

Interest 12.3 3.5 26 58.4

Intelligence 2.7 6.2 17.8 12.4

Comprehension - 3.5 21.9 7.1

Reflection 28.8 13.3 1.4 5.3

Engagement 6 45 16.4 8.8

TABLE 4. Percentages of answers to the question about the importance/unimportance of the skills in both the groups (MG and LG)

Considering the whole sample, one of the most indicated skills is concen-tration which assumes a different degree of importance in the two groups (mathematics vs. language): it is the most important element according to the 45.2% of the mathematics group (MG) and the 25.7% of the language group (LG). Engagement was judged in a completely different way in the two groups: it was important for the 45% of the LG an only for the 6% of the MG. The LG maintained the unimportance of the interest for the matter (58.4%) and in-telligence (12.4%), while the MG said that reflection is an important skill (28.8%), but this was not the case for comprehension (21.9%).

By crosstabulating responses given to the request to select the most and to the request to select the least important alternative (Table 5 and 6), it emerged that in LG, children who believed that engagement, reflection or concentration are the most important skills, said that the least important one is to be interested in the matter (cumulative percentage among those children who chose interest: 83.4%; X2(N = 113) = 76.782, df = 36, p < .001). In this group, the choice of interest in the school matter as an unimportant skill was also related to the choice of gifts and challenge as unimportant motivations (X2(N = 113) = 50.806, df = 24, p = .001) and of questioning the teacher, making exercises and

Manuela Cantoia

106

ask for an help as unimportant learning behaviours (X2(N = 110) = 38.922, df = 18, p < .005). Data about the relation between important skills and learning behaviour tended to be significant: when engagement, concentration and reflection were pointed out as important skills, children said that the most important learning behaviour is attention (X2(N = 113) = 28.193, df = 18, p = .059).

Importantskill

Not important skill

Language Concen-tration Memory Interest Intelli-

genceCompre-hension

Reflec-tion

Engage-ment

Concentration - 1 14 3 2 4 4

Memory - - 4 - - 1 1

Interest - - - 4 - - -

Intelligence 1 1 3 - 2 - -

Comprehension - - 4 - - - -

Reflection - 1 11 1 2 - -

Engagement 1 4 30 6 3 1 1

TABLE 5. Crosstabulation between the important/unimportant skills in the LG

Important skill

Not important skill

Mathematics Concen-tration Memory Interest I n t e l l i -

genceCompre -hension

Ref lec -tion

E n g a -gement

Concentration - 3 8 7 7 - 8

Memory 1 - 1 - - - -

Interest - 3 - - 3 1 2

Intelligence - - - - - - 2

Reflection - 5 9 3 4 - -

Engagement - - 1 3 2 - -

TABLE 6. Crosstabulation between the important/unimportant skills in the MG

In MG, when concentration was a good skill, then interest (24.2%), engagement (24.2%), intelligence (21.2%), comprehension (21.2%) and memory (9.1%) were not; but when children choose reflection as the important skill, then they said that interest (42.9%), memory (23.8%), comprehension (19%) and intelligence (14.3%) are


107

not important skills (X2(N = 73) = 76.481, df = 30, p < .001). Still, when chil-dren pointed out the importance of concentration and reflection and the unim-portance of memory, interest, intelligence, comprehension or engagement, they tended to think that gifts and challenge are not good motivations (respectively impor-tant skill: X2(N = 73) = 42.436, df = 20, p < .01; unimportant skill: X2(N = 73) = 45.815, df = 24, p = .005), that revising and reflection are important learn-ing goals (this datum just tends to be significant for the important skill: X2(N = 72) = 37.522, df = 25, p = .051; unimportant skill: X2(N = 72) = 47.851, df = 30, p < .05), and that to ask for an help and questioning the teacher are not important learning behaviours (important skill: X2(N = 73) = 25.657, df = 15, p < .05; unimportant skill: X2(N = 73) = 2.575, df = 18, p < .05).

Children who believe that concentration is an important skill tend to reject external attributions (e.g. to be lucky/unlucky, too short time, difficultness of the task; X2(N = 73) = 62.092 df=40 p<.05). In the MG, the trend of answers according to the variable of school performance (Table 7) was very interest-ing (X2(N = 72) = 41.700, df = 24, p < .05): memory is chosen as a not impor-tant skill from those students who have a low performance, those who have medium performances refer to interest, and children with high performances say engagement or comprehension.

Not important skill in the MG

School PerformanceNot sufficient Sufficient Good Very good Optimum

Concentration 12.5 - - - -Memory 37.5 27.8 9.1 7.7 -Interest 12.5 27.8 36.4 23.1 18.2Intelligence - 22.2 18.2 7.7 27.3Comprehension 12.5 22.2 18.2 15.4 45.5Reflection 12.5 - - - -Engagement 12.5 - 18.2 46.2 9.1

TABLE 7. Distribution (percentages) of the unimportant skills in the MG according to school performance

Does these data allow to hypothesize a low metacognitive level, particu-larly in children who have high performances?

In general, in the MG females seem to have clearer ideas about the im-portant skills, because they choose less elements but in a larger percentage, as

Manuela Cantoia

108

in the case of concentration (Female 54.1%, Male 36.1%) and reflection (Female 35.1%, Male 22.1%; X2(N = 73) = 12.775, df = 5, p < .05) (Table 8).

Important skill in the MG GenderMale Female

Concentration 36.1 54.1Memory 5.6 -Interest 13.9 10.8Intelligence 5.6 -Reflection 22.1 35.1Engagement 16.7 -

Comprehension was not chosen in the MG.

TABLE 8. Distribution (percentages) of the important skills in the MG according to gender

Data analysis according to grade in the MG (X2(N = 73) = 19.390, df = 10, p < .05) showed that the importance of concentration decreased in the IV grade (65.4% in the III grade, 22.7% in the IV grade, 44% in the V grade), where reflection increased instead (respectively 11.5%, 40.9%, 36%); the im-portance of engagement increased from the III to the V grade (respectively 3.8%, 9.1%, 12%), but the interest decreased (respectively 11.5%, 18.2%, 8%). As far as the unimportant skills were concerned (X2(N = 73) = 23.236, df = 12, p < .05): memory and intelligence increased by the age (memory: 3.8%, 18.2%, 24%; intelligence: 11.5%, 13.6%, 28%), comprehension and interest de-creased (comprehension: 30.8%, 31.8%, 4%; interest: 19.2%, 31.8%, 28%), while engagement is not mentioned in the IV grade but in the other two grades (respectively 30.8%, 0%, 16%).

Learning goalsWhen asked to choose the most important learning goal among seven

possible choices (to read and repeat in someway, to comprehend, to revise, to resume, to memorise, to reflect, to make maps or schemes), children in the LG answered prevalently “to comprehend” (51.3%), while children in the MG preferred “to revise” (39.2%). The second choice of both groups was “to reflect” (respectively: LG 30.1%; MG 33.8%; X2(N = 187) = 72,035, df = 6, p < .001) (Table 9). Data did not show any significant difference between the


109

two groups according to the least important learning goal (X2(N = 188) = 8.770, df = 6, p = .187).

Learning goalsImportant goal

MG LGTo read/repeat in someway 1.4 3.5To comprehend - 51.3Revising 39.2 8.8To make synthesis /Resume 10.8 3.5Memorising 10.8 2.7To reflect 33.8 30.1To make maps/scheme 4.1 -

TABLE 9. Distribution (percentages) of the most important learning goals

In the MG answers differed according to the school performance variable (X2(N = 73) = 48.442, df = 20, p < .001): children with low performances said that memorising is an important goal, children with medium performances pre-ferred revising and those with high performances said that is important to reflect. The importance of revising and memory decreased by the age (revising: 51.9% in the III grade, 31.8% in the IV grade, 32% in the V grade; memory:14.8%, 13.6%, 4%), reflection grows (14.8%, 27.3%, 60%); “to resume” was high only in the IV grade (7.4%, 22.7%, 4%) (X2(N = 74) = 23.450, df = 10, p < .01).

Important goal in the MG School performance

Not sufficient

Sufficient Good Very good Optimum

To read/repeat in someway - - - - 9.1Revising 22.2 50 45.5 46.2 18.2Make synthesis /Resume - 16.7 13.6 7.7 9.1Memorising 44.4 5.6 4.5 - 18.2Reflect - 27.8 36.4 46.2 45.5Make maps/scheme 33.3 - - - -

Comprehension was not chosen in the MG

TABLE 10. Distribution (percentages) of the most important learning goals according to school per-formance variable in the MG

In this same group, when children said that revising and reflection are impor-tant, they also believed that attention was an important learning behaviour and

Manuela Cantoia

110

secondarily questioning the teacher and making exercises (X2(N = 74) 28.563 df = 15, p < .05). Talking about attribution, when children chose either revising or reflection, they rejected external attribution (difficultness and too short time: X2(N = 74) = 56.815, df = 40, p < .05), but they showed different opinions about the important attribution: children who chose revising pointed out the importance of studying and secondarily of keep on trying, while children who preferred reflection chose more frequently the importance of keep on trying and then of studying (X2(N = 74) = 66.284, df = 35, p = .001). External attri-bution was rejected also by children who thought that memorising and making maps/schemes or to study in someway were not good learning goals (X2(N = 75) = 73.316, df = 48, p = .011).

In the LG a significant relation between important and not important learning goals emerged: children who said that comprehension and reflection are important also thought that it is not worth studying someway, memorising or mak-ing maps/schemes (X2(N = 113) = 49.526, df = 30, p < .05). The choice of the least important goals was related also to the choice of the least important strategy: children who rejected studying someway, memorising or making maps/schemes agreed on memorising and making examples as not important strategies, but making attempts was rejected by those who had chose memorising, while the second choice of those who refused studying someway was to make analogies with previous situations (X2(N=113) = 76.877, df = 48, p = .005). This choice was also related to the least important attribution (X2(N=113) = 66.952, df = 48, p < .05) and to the most important one (X2(N=113) = 88.980, df = 48, p < .001: when studying someway, memorising and making maps/schemes were not important learning goals, then time/noises and admit that one has not study were not good attributions; on the contrary, to have studied and to keep on trying were good.

The choice of the most important learning goal was related to the most important attribution (X2(N=113) = 66.226, df = 40, p < .01: when revising and reflecting were good goals, then studying and keep on trying are the most quoted attributions, even if some children accepted also an external attribu-tion as to be lucky.

Data analysis according to grade in the LG showed that memory, as not important goals, increased by the age (respectively 2.7% in the III grade, 37.8% in the IV grade and 33.3% in the V grade), but to study someway and


111

to make maps decreased in the IV grade (study: 32.4%, 27%, 38.5%; maps: 29.7%, 13.5%, 28.2%). It is worth notice that in the V grade children never mentioned comprehension, revising, to resume and reflection as unimportant goals.

StrategiesChildren could choose among nine alternatives (memorising, to put to-

gether/to break down, visualising, temporal sequence, cause, attempt, mak-ing examples, to categorise, analogy). Children’s answers were very different both between and within groups, as in the case of memorising: in the MG the 25.3% of the children said it is a good strategy, but in the LG the 26.3% said that it is not; but for the 20% of the MG memorising was not important and for the 19.6% in the LG it was (see Table 11) (the most important: X2(N = 188) = 49.236, df = 8, p < .001; the least important: X2(N = 188) = 34.546, df = 8, p < .001).

StrategyImportant Strategy Unimportant Strategy

MG LG MG LGMemorising 25.3 19.5 20 26.5To put together/to break down 1.3 21.2 4 6.2Visualising 6.7 17.7 24 7.1

Temporal sequence - 3.5 14.7 3.5

Cause 8 3.5 1.3 2.7

Attempt 16 19.5 5.3 7.1

Making examples 33.3 3.5 - 21.2

To categorise 2.7 4.4 16 12.4

Analogy 6.7 7.1 14.7 13.3

TABLE 11. Distribution (percentages) of the more/less important strategies

An unexpected trend emerged for the strategy of visualising. One could expect that it is a strategy more suitable for mathematics, but children in the MG made a different choice: for the 17.7% of the LG it was important, while for the 24% of the MG it was not important. The same expectations are disappointed in the case of the importance of the attempts, which were the second choice (19.5%) in the LG and the third (16%) in the MG. Students

Manuela Cantoia

112

tended to neglect those strategies that generally teachers use to stress, for example categorising, making temporal sequences, find causal links.

The answers of both groups were significantly different in males and fe-males (Table 12): in the MG (X2(N = 75) = 16.054, df = 7, p < .05) girls said that memorising is not an important strategy, while boys preferred visualising. In the LG (X2(N = 113) = 16.138, df = 8, p < .05) the two genders agreed about making examples and memorising, even if more males chose this latter one.

Not important strategyMG LG

Male Female Male FemaleMemorising 10.5 29.7 30 22.6To put together/to break down 2.6 5.4 - 13.2

Visualising 34.2 13.5 5 9.4Temporal sequence 18.4 10.8 6.7 -Cause 2.6 - 3.3 1.9Attempt 10.5 - 8.3 5.7Making examples * * 21.7 20.8To categorise 7.9 24.3 8.3 17Analogy 13.2 16.2 16.7 9.4

* This item was not chosen in the MG.

TABLE 12. Distribution (percentages) of the unimportant strategies in the two groups according to gender

In the LG the choices of the least/most important strategy were signifi-cantly related: if memorising was important, making examples was not; if visualis-ing and to put together/break down were important, then memorising and making examples were not; when visualising and making attempts were important, to categorise was not important as well as making analogies with previous situations (X2(N = 113) = 99.249, df = 64, p < .005).

According to grade, in the LG the importance (X2(N = 113) = 38.015, df = 16, p < .005) of memory (40.5% in the III grade, 16.2% in the IV grade, 2.6% in the V grade), analogy (8.1%, 10.8%, 2.6%) and attempts (21.6%, 24.3%, 12.8%) decreased by the age, while visualising (8.1%, 24.3%, 20.5%) and to put together/break down (10.8%, 13.5%, 38.5%) increased. The least important strategy (X2(N = 113) = 39.886, df = 16, p = .001) was more and more memory (16.2%, 16.2%, 46.2%), while visualisation decreased (13.5%, 2.7%, 5.1%). In the IV grade making examples decreased (32.4%, 10.8%,


113

20.5%) and categorising increased (5.4%, 27%, 5.1%). Also in the MG mem-ory decreased its importance as good strategy (X2(N=75) = 31.246, df = 14, p = .005; respectively 59.3%, 4.1%, 3.8%), while temporal causation (3.7%, 4.5%, 15.4%), making examples (18.5%, 45.5%, 38.5%) and the attempts (7.4%, 18.2%, 23.1%) increased.

AttributionChildren had to choose between nine possibilities (to be lucky/unlucky, to

study/not to study, difficult/easy task, external help, time/noises, to keep on trying). Both groups agreed about either the important or the unimportant attribution: it was important (X2(N=188) = 68.560, df = 8, p < .001) to keep on trying (MG 37.3%, LG 42.5%) and to study, particularly for the MG (MG 36%, LG 23.9%); it was not important (X2(N=188) = 30.845, df = 8, p < .001) to justify oneself by an external attribution as time or noises (MG 26.7%; LG 32.7%) (Table 13).

AttributionImportant Attribution Unimportant Attribution

MG LG MG LGFortune - 15.9 12 10.6

To study 36 23.9 4 4.4Not to study 5.3 7.1 5.3 23Bad luck 2.7 9 12 7.1Easy 5.3 2.7 6.7 7.1Difficult 4 1.8 16 7.1I’ll ask for an help 5.3 2.7 16 7.1I’ll keep on trying 37.3 42.5 1.3 .9Lack of time/too much noises 4 2.7 26.7 32.7

TABLE 13. Distribution (percentages) of the more/less important attributions in the two groups

It is worth showing that for the 23% of the LG it was not important to admit that one has not studied: who are those children? Since in this group the choices of the most/least important attribution were significantly related, we can see that 33.3% of the children who did not recognise the importance of admitting that one has not studied said that the most important attribution is to keep on trying, studying and fortune. The children who said that time is not

Manuela Cantoia

114

important recognised instead the importance of keep on trying and studying. For the most part of children who said that fortune is not important, keep on trying was a good attribution (X2(N = 113) = 173.724, df = 64, p < .001).

In the MG responses were associated to school performance (X2(N=74) = 47.922, df = 32, p < .05): children with low performances said that to admit that one had not studied is not important; children who have medium perform-ances recognised that external attribution are not important as well as children with high performances, but they differed in the typology of attribution they mentioned: the first ones pointed out time and noises, while high performance pupils selected bad luck or difficultness of the task. In the same group, children who thought that keep on trying is a good attribution rejected memorising, visu-alising and making categories as good strategies; if children choose the impor-tance of studying instead, they refused visualising and making analogies as good strategies (X2(N=75) = 67.968, df = 49, p < .05) (Table 14).

Not important attribution in the MG

School performanceNot

sufficientSufficient Good Very good Optimum

Fortune 22.2 5.3 9.1 15.4 18.2To study 11.1 5.3 4.5 - -Not to study 33.3 5.3 - - -Bad luck 11.1 5.3 9.1 30.8 9.1

Easy - 15.8 4.5 - 9.1

Difficult - 10.5 18.2 15.4 36.4

I’ll ask for an help 11.1 15.8 13.6 23.1 18.2

I’ll keep on trying 11.1 - - - -

Lack of time/too much noises - 36.8 40.9 15.4 9.1

TABLE 14. Distribution (percentages) of the less important attributions according to school per-formance in the MG

As far as grade was concerned, in the MG the importance of studying and to keep on trying showed different trends: studying decreased (44.4% in the III grade, 45.5% in the IV grade, 19.2 in the V grade), and the other alterna-tive increased (7.4%, 45.5%, 61.5%). It is worth noticing that in the IV grade children never mentioned external attributions like to be unlucky, easiness of the task, time/noises and even not the request for a help (X2(N = 75) =


115

27.492, df = 14, p < .05). In the LG, on the contrary, fortune decreased its importance (35.5%, 13.5%, 0%), while to study (16.2%,18.9%, 35.9%), to keep on trying (32.4%, 51.4%, 43.6%) and time/noises (0%, 2.7%, 5.1%) increased (X2(N = 113) = 27.235, df = 16, p < .05).

MotivationChildren could choose between five possibilities (gift/punishment, duty,

pleasure, satisfaction, challenge). By crosstabulating the choices of the al-ternatives and the subject matter (MG vs. LG) the following values were obtained: important motivation (X2(N = 188) = 45.583, df = 4, p < .001; not important motivation: (X2(N = 188) = 9.503, df = 4, p = .050. Pupils generally agreed about the importance of the sense of duty (MG 46.7%, LG 40.7%), but they differed about the importance of looking for personal satis-faction that was preferred by the MG (21.3%) or pleasure (personal interest), preferred by the LG (48.7%). Children in the MG refused more decidedly the importance of receiving gifts/punishments (MG 60%, LG 46%), while in the LG children maintained the more that the challenge is not important (MG 21.3%, LG 33.6%) (Table 15).

MotivationImportant motivation Unimportant motivation

MG LG MG LG

To receive a gift/avoid a punishment 17.3 7.1 60 46

Duty 46.7 40.7 2.7 8

Pleasure 9.3 48.7 4 7.1

Personal satisfaction 21.3 1.8 12 5.3

Challenge 5.3 1.8 21.3 33.6

TABLE 15. Distribution (percentages) of the important/not important motivations

In the LG there was a significant relation between the two choices: chil-dren who said that the most important motivation are duty or pleasure, rec-ognised that gifts were not important as well as the challenge (X2(N = 113) = 39.314, df = 16, p = .001). This choice about the least important motivation was also related to memorising, making maps or schemes and studying someway as the least important learning goals (X2 (N = 113) = 54.014, df = 24, p < .001)

Manuela Cantoia

116

and to attention as the most important learning behaviour (X2 (N = 113) = 22.449, df = 12, p < .05). The choice of duty and pleasure were related to keep on trying as the most important attribution, but while the second choice of the children who prefer duty was to be lucky, those children who pointed out pleasure chose also studying (X2 (N = 113) = 66.785, df = 32, p < .001). Im-portant motivations only tended to be related to the choice of comprehension and reflection as the most important learning goals (X2 (N = 113) = 30.536, df = 20, p = .062).

In the LG the importance of duty decreased by grade (62.2% in the III grade, 43.2% in the IV grade, 17.9% in the V grade), and that one of pleasure increased (29.7%, 40.5%, 74.4%; X2 (N = 113) = 23.889, df = 8, p < .005). The importance of receiving a gift/avoid a punishment increased in the IV grade (5.4%, 10.8%, 5.1%). In the MG data about the most important mo-tivation were significant associated to gender (X2 (N = 75) = 10.301, df = 4, p < .05) (Table 16), while data about the least important motivation were significantly associated to school performance (X2 (N = 74) = 29.075, df = 16, p < .05) (Table 17). Boys and girls in the MG agreed about the importance of the duty, even if in a different degree (respectively, 42.1% and 51.4%), but they showed a different second choice: males said that gifts and punishments are good motivation; females preferred personal satisfaction instead. Low per-formance children affirmed that personal satisfaction is not important, while the choice of gift/punishment as not important motivation increased with the performance level. All the children agreed about the challenge, which was not considered an important motivation.

Important motivation in the MGGender

Male Female

To receive a gift/avoid a punishment 26.3 8.1

Duty 42.1 51.4Pleasure 5.3 13.5Personal satisfaction 15.8 27Challenge 10.5 -

TABLE 16. Distribution (percentages) of the important motivations in the MG according to gender


117

Not important motivation in the MG

School performanceNot

sufficient Sufficient Good Very good Optimum

To receive a gift/avoid a punish-ment 11.1 68.4 54.5 76.9 72.7

Duty - 5.3 4.5 - -Pleasure 11.1 - 4.5 - 9.1Personal satisfaction 55.6 5.3 4.5 4.5 -Challenge 22.2 21.1 31.8 31.8 18.2

TABLE 17. Distribution (percentages) of the not important motivations in the MG according to school performance

In the same group the choice of the least/most important motivation were related each other (X2 (N = 75) = 34.025, df = 16, p = .005): children who were not motivated by gifts or challenge, were motivated by the sense of duty and satisfaction and they also thought that external attribution (time/noises, difficultness, to be lucky/unlucky) is not a good one (X2 (N = 75) = 67.046, df = 32, p < .001).

Learning behaviourChildren could choose among four alternatives (attention, help, questions,

exercises). Crosstabulation of the choices of these alternatives and subject matter

produced the following values: important learning behaviour X2 (N = 188) = 13.467, df = 3, p < .005; not important learning behaviour X2 (N = 188) = 8.270, df = 3, p < .05 (Table 17).

Children in both groups said that attention is the most important learning behaviour, even if in a different percentage (MG; 45.3%; LG: 63.7%); they also agreed about the least important behaviour: it is not important to question the teacher (MG: 46.7%; LG: 37.3%) and to ask for a help (MG: 36%; LG 36.4%) (Table 18).

Manuela Cantoia

118

Learning behaviourImportant behaviour Not important behaviour

MG LG MG LG

Attention 45.3 63.7 6.7 1.8

Ask for an help 10.7 16.8 36 36.4

Questioning the teacher 22.7 11.5 46.7 37.3

Exercises 21.3 8 10.7 24.5

TABLE 18. Distribution (percentages) of important/not important learning behaviours

In the MG the choice of the least important learning behaviour was relat-ed to the most important strategy (X2 (N = 75) = 44.598, df = 21, p < .005): if asking for an help and questioning the teacher were not important, then memo-rising, making attempts or exercises were good strategies. In this group motiva-tion and learning behaviour were related at different levels: children who said that attention is an important learning behaviour were motivated by the sense of duty (50%), the satisfaction (23.5%), but also by the gifts (20.6%); children who chose questioning and making exercises as important learning behaviours were motivated above all by the sense of duty (X2 (N = 75) = 21.600, df = 12, p < .05); children who made one of these choices (attention, questioning, mak-ing exercises) about the important behaviour were not motivated by gifts and challenge (X2 (N = 75) = 28.691, df = 12, p < .005). On the contrary, children who rejected asking for help and questioning were not motivated neither by gifts nor by challenge (X2 (N = 75) = 25.446, df = 12, p < .05). According to grade, in this group the importance (X2 (N = 75) = 14.565, df = 6, p < .05) of the attention decreased in the IV grade (respectively 55.6% in the III grade, 18.2% in the IV grade, 57.7% in the V grade), but in the case of asking for a help and exercises it increased (first one: 11.1%, 22.7%, 0%; second one: 11.1%,36.4%, 19.2%). In the LG important and unimportant learning behaviours were re-lated each other (X2 (N = 110) = 38.212, df = 9, p < .001): children who said that attention is important thought that asking for an help (46.4%), questioning (34.8%) and making exercises (18.8%) are not important. Children who pre-ferred to make questions to the teacher, choose neither making exercises (61.5%) nor asking for help (30.8%) and children who asked for help neither made ques-tions to the teacher (68.4%) nor made exercises (31.6%).

Table 19 and 20 summarise the significant relations among the answers to the issues that have been discussed since now.


119

TABLE 19. Relations between the answers to the different fields of analysis in the MG

TABLE 20. Relations between the answers to the different fields of analysis in the LG

Visualising/analogy (S-)

Memory/visualising/categorising(S-)

To study (A+) Keep on trying (A+)

Memory/maps/ to study someway (LG-)

Revising/reflection (LG+)

Questioning/Help (LB-)

Memory/attention/example (S+)

Gift/challenge(M-)

Duty/satisfaction(M+)

External attribution(A-)

Attention/questioning/exercices (LB+)

Concentration/reflection(S+)

Interest/engagement/intelligence/comprehen-sion (S-)

Mathematics groupM = motivation; LG = learning goal; LB = learning behaviour; SK = skill; S = strategy; A = attribution

Gift/challenge(M-)

Interest (S-)

Attention (LB+)

To study someway/memory/maps (LG-)

Duty/pleasure (M+)

Questioning/exercise/help (LB-)

Engagement/concentration/reflection (S+)

Memory/example (S-)

Comprehension/reflection (LG+)

No not study (A-)Duty/pleasure (M+)

To study someway/memory/maps (LG-)(no ext.attr)

To study/to keep on trying/external attribution (A+)

Memory (S+)Example (S-) Comprehension/reflection (LG+)

No not study/external attribution (A-)

Language groupM = motivation; LG = learning goal; LB = learning behaviour; SK = skill; S = strategy; A = attribution

Manuela Cantoia

120

DISCUSSION

This study was aimed at finding differences in the representations of a few basic elements (skill, goals, strategies, attribution, motivation and learning be-haviour) in two different domains, mathematics and language. I will start this discussion by trying to present a clearer picture of the results.

First, differences between genders or among different grade and levels of performance emerged more often in the MG than in the LG. Maybe this is a confirmation of the degree of complexity of the representations of math-ematics and of the general critical role of the IV grade (Lucangeli, 1995).

Particularly, responses about the important motivation diverged from the data in literature about children’s belief that mathematics is a typical male matter (Lucangeli, 1995; Poli & Zan, 1996): in this sample girls choose per-sonal satisfaction while boys preferred to receive gift/avoid punishment (both genders mentioned also the sense of the duty). Answers about motivation were also influenced by the performance level: gift/punishment as not impor-tant motivation increased with the performance level, and only low perform-ance children affirmed that personal satisfaction was not important. In the LG, this choice was influenced by a developmental trend: the importance of duty decreased by grade and that one of pleasure increased. Only in the IV grade the importance of receiving a gift/avoid a punishment increased.

Nevertheless, as far as attribution is concerned, in the MG children in the V grade compared to the pupils in the other two grades, thought that study-ing is less and less important, but they gave more importance to the personal engagement (to keep on trying). On the contrary, children in the LG in the V grade believed that to study is an important attribution and they pointed out also the increasing importance of an external attribution as time/noises, that is to say the element that can affect concentration (in fact they think that for-tune is not important).

Children thought that to be interested in the matter is not important for studying language (but they said that the most important learning goal is to comprehend), while comprehension, engagement and memory were not useful prereq-uisites in mathematics. Making attempts was chosen as a good strategy by both groups (only for mathematics it is worth using examples to understand in a better way).


121

At times data seem to show that in someway students are more and more personally concerned by the matter as they grow or as they succeed their performances, but on the other hand, in other situations, children do not recognise the value of certain important skills or of studying (as in the case of the MG in the important attribution).

Maybe this ambivalent picture is the result of different levels of metacog-nitive competence in the children of the sample, or maybe there was not a complete agreement on the sense and value of the terms we presented, that is to say, in the experience of each class some element could be more or less known, employed in the everyday speech or stressed by the teachers. The gap between the representations of the adults and those of the pupils, the influ-ence of the metacognitive competence and the habits of the single classes are somehow confirmed by a few other elements.

Children often chose items that are very general or taken from the adult world speech (many issues are typically evoked by the teachers in the class-room). This is the case of concentration, reflection and engagement as important skills; reflection as the important learning goal; attention as important learning behaviour; sense of duty as important motivation and in a certain sense may-be also the choice of personal satisfaction and personal pleasure which are ele-ments of intrinsic motivation whose importance is often recalled by adults, as well as in the case of gift/punishment and challenge as unimportant motivations.

Children thought that questioning the teacher or ask for a help are not im-portant behaviours in situations of difficultness. We already knew about the conviction that one has to succeed by himself in mathematics (Poli & Zan, 1996), but since the same answer was given by the LG, one possible explana-tion could be found at the relational level: do children feel and know the avail-ability of their teacher in problematic situations? Do their teachers behave with an open stance towards the difficulties of their class?

Maybe the conviction about the unimportance of the external helps comes from the attitude of the teachers rather than from a belief based on the char-acteristics of a particular school matter.

Up to now, the discussion has provided arguments about the relation and the reciprocal influence between children and adults. In a previous study (Can-toia, 2002) the answers (which in such investigation referred only to math-ematics) of the children were compared with those of their teachers (the two

Manuela Cantoia

122

parts were also asked to preview the other’s answers). The main result of that study was to deepen the difference between child and adult conception on the same argument. In many situations pupils answered in a different way than their teacher, but the more interesting result was that children could bet-ter preview their teacher’s expectations than the contrary (teachers were often sure about certain alternatives that pupils had not chosen).

In this study, the picture that emerges from the data seems to underline the importance of considering the influence of the relation with the teacher in the construction of a learning conception: the model given by the teacher, his/her styles of thinking and teaching, his/her way of making sense of the school matter itself, his/her way to present and organise the lessons (Mason & Scrivani, 2004) take the student to a particular conception both of the mat-ter and of what one has to do to learn it (Sternberg, 1997).

The second argument of this discussion concerns the issues of strate-gies and, more generally, of metacognition. When asked about the strategies, children seemed to be confused about the role of memory, which appears both as important and unimportant element in the two groups. The ques-tion of memory is relevant because at school children are exposed both to positive and negative arguments about memory: in certain situations they are asked to learn by heart, but in other moments they are told that is not good to memorise the lesson, because they should revising texts and comprehend the meaning.

Still, in a few tasks, children pointed out choices that one could expect in the opposite field of study, as in the case of the choice of revising as important learning goal for mathematics or visualising as important strategy for language. So, are children really able to distinguish the more suitable use for one same strategy?

This question introduces the last issue. Let me take an example: when asked about attribution, children in this sample seemed to be sure about the importance of internal attribution (to study, to keep on trying) and this datum could be confirmed by the parallel rejection of external attributions (noises, difficultness, etc.). Nevertheless, particularly in the LG, a few children choose elements of both the types and, what is more important, they did not seem to recognise the importance of admitting that one has not studied. This situation arises a core question: do those children think that this admission is rather


123

like an excuse or have they a low metacognitive level which does not allow them to find out the real cause of their failure? This question recurs twice in the MG: children at all the levels of performance said that admitting that one has not studied is not an important attribution; then children with very high performance said that comprehension is not an important skill.

Data from the two groups also showed that the level of consciousness about the real value of certain items often increased with the age, so a few critic answers can be explained by a developmental trend.

From the latest issues one maybe derive an argument about the influence that the degree of metacognitive competence has on the making of learn-ing conceptions of the single school matters. Sometimes the opportunity or the difficultness to understand the real value of certain elements (strategies, behaviours, and so on) and their proper use lead the student to false beliefs or to false generalizations that affect both his representations of the school matter and of the elements which are related to it.

This study was aimed at probing the representations of two different fields of instruction with a global approach, that is to say, trying to deepen a set of issues that converge in the representation-making (attribution, motiva-tion, goals, behaviour, skill and strategy). The results showed both similarities and differences that have been explained according to the relational level and the metacognitive competence. Maybe it is useful at this point to distinguish two different conclusions: the administration itself represented an important opportunity for teachers and children to face the issue of learning concep-tions. They admitted they had never had before the occasion to take a time to reflect about these arguments and to deepen their consciousness about them-selves and about such arguments. In this sense, the answers of the children in this sample were given without a familiarity with such arguments and without prior opportunities for reasoning about them. After the administration each class had an informal discussion about the experience and the children had the possibility to match their answers and their beliefs and to make some ele-ment clearer.

This study confirmed the well-constructed and complex nature of learn-ing conceptions: they are the result of many core elements. As many studies in literature already showed, data in this contribution give a picture of the representation of mathematics (which is a much more probed field) as more

Manuela Cantoia

124

influenced by gender, grade, and school performance. Anyway, the differ-ences in the representations of the two domains are affected by many im-portant matters as the schooling experiences, the relation with the teacher, the teaching style of the teacher and his pedagogical choices, the degree of comprehension and consciousness of all the elements involved.

REFERENCES

Antonietti A. (1995, ed.), Il divenire del pensiero, Raffaello Cortina, MilanAntonietti A. (1998), Psicologia dell’apprendimento, La Scuola, BresciaAntonietti A., Cantoia M. (2000), La mente che impara, La Nuova Italia, Mi-

lanCampbell J., Smith D., Boulton –Lewis G., Brownlee J., Burnett P.C., Car-

rington S. and Purdie N. (2001), Students’ perceptions of teaching and learning: the influence of students’approaches to learning and teachers’ approaches to teaching. Teachers and Teaching: theory and practice, vol. 7 (2):173-187

Cantoia M. (1999), La Matematica per me è.., Scuola Italiana Moderna, 11Cantoia M. (1998), Metacognizione in classe: strumento per la riflessione e di-

scussione in classe, Congresso Nazionale della Sezione di Psicologia dello Sviluppo dell’AIP, Bressanone, 5-7 December 1998

Cantoia M. (2000), “Leggere” la mente dell’insegnante e la propria: la com-prensione di sé e dell’altro nella dinamica dell’apprendimento, in Marchetti A. (a cura di), Incontri evolutivi: crescere nei contesti attraverso le relazioni, Franco Angeli, Milan

Cantoia M. (2000), Intorno alla matematica: credenze di alunni ed insegnanti del secondo ciclo elementare, Convegno nazionale SIPEF, Naples, Sep-tember 2000

Cantoia M. (2002), Matematica e metacognizione: alunni ed insegnanti della scuola elementare a confronto, Convegno Nazionale Sipef, Urbino De-cember 2002

Cantoia M. (2003), Approccio all’apprendimento nella scuola elementare: Matematica e Lingua a confronto, Convegno Contesto, cultura, interven-to. Quale psicologia per la scuola del futuro, Lecce, 20-22 June 2003


125

Cantoia M., Antonietti A. (1997), Il buon solutore di problemi nella rappre-sentazione degli studenti universitari, Studi di Psicologia dell’Educazione, 1-2

Cantoia M., Di Tullio F., Ornaghi V., Antonietti A. (1999), Comportamenti di studio in alunni di scuola elementare e media, Psicologia e Scuola, ott-nov., p.3-12

Cornoldi C. (1995), Metacognizione e apprendimento, Il Mulino, BolognaEntwistle N., and Marton F. (1994), Knowledge objects: understandings con-

stituted through intensive accademic study, British Journal of Educational Psychology, 64: 161-178

Gibbs G. (1995), Changing lecturers’ conceptions of learning through ac-tion research. In A. Brew (Ed.), Directions in staff development. Buck-ingham, Society for Research in Higher Education and Open University Press, GB

Gow L., Kember D. (1993), Conceptions of teaching and their relationship to student learning, British Journal Educational Psychology, 63: 20-33

Klatter E. B., Lodewijks H. and Arnoutse C. (2001), Learning conceptions of young students in the final year of primary education, Learning and Instruction, December 11 (6): 485-516

Lucangeli D., Passolunghi M.C. (1995), Psicologia dell’apprendimento mate-matico, UTET, Torino

Martin E., Ramsden P. (1987), Learning skills or skills in learning? In: JTE Richardson; M.W. Eysenck (eds.) Student learning research in education and cognitive psychology. Open University Press, Milton Keynes

Marton F. (1988), Describing and improving learning, in R.R. Schmeck (ed.) Learning Strategies and Learning Styles (Chapter 3) Plenum, New York

Mason L. (2001) Verità e certezze. Natura e sviluppo delle epistemologie in-genue, Carocci, Rome

Mason L., Scrivani L. (2004), Enhancing student’s mathematical beliefs: an intervention study, Learning and Istruction, 14, 153-176

Mclean M. (2001), Can we relate conceptions of learning to student academic achievement?, Teaching in Higher Education, 6, 3

Mezirow J. (1991), Trasformative dimensions of adult learning, Jossey-Bass, San Francisco, CA

Pellerey M. (1996), La dimensione affettiva e motivazionale nei processi di apprendimento della matematica, Studi di Psicologia dell’educazione, 15

Manuela Cantoia

126

(1-2), 19-36Pellerey M., Orio F. (1995), La diagnosi delle strategie cognitive, affettive,

motivazionali coinvolte nell’apprendimento scolastico, Orientamenti Pe-dagogici, 4, 683-726

Poli P., Zan R. (1996), Il ruolo delle convinzioni nella risoluzione dei proble-mi, La matematica e la sua didattica, 4, 440-466

Pontecorvo C. (1999), Manuale di Psicologia dell’educazione, Il Mulino, Bo-logna

Pramling I. (1988), Developing children’s’ thinking about their own learning, British Journal Educational Psychology, 58, 266-278

Säljö R. (1979), Learning in the learners’ perspectives 1. some commonsense conceptions, Reports from the Department of Education, University of Gothenburg, 76

Sternberg R. (1997), Stili di pensiero. Differenze individuali nell’apprendi-mento e nella soluzione di problemi, Erickson, Trento 1998

Van Rossum Ej, Schenk Sm. (1984), The relationship between learning con-ception, study, strategy and learning outcome, British Journal of Educa-tional Psychology, 54, 73-83

Zan R. (1998), Problemi e convinzioni, Pitagora, Bologna


127

CONCEPTIONS OF LEARNING AND USE OF CULTURAL MEDIA

Sonia Pérez-Tello*, Alessandro Antonietti**, Olga Liverta-Sempio**, Antonella Marchetti***

Relationships between folk conceptions of learning and media use were investi-gated. Three hundred and thirty-five secondary school students were administered a questionnaire aimed at assessing their beliefs about cognitive, emotional, and at-tributive aspects of learning and a questionnaire measuring the frequency of use of television, videorecorder, audio-cassette and CD player, radio, computer, books, newspapers and journals, as well as museum, exhibition, theatre, concert and cinema attendance. Analyses highlighted the multidimensional nature of the conceptions of learning and showed significant links between some aspects of such conceptions and the frequent use of specific media.

INTRODUCTION

The studies about the naive conceptions of learning (Marton & Säljö, 1976) – that is, the convictions that students develop about the process of learning – investigated relationships with aspects such as school results (Lon-ka & Lindblom, 1986), study approaches (Van Rossum & Schenk, 1984), self-regulation (Slaats, Loderwijks & Van der Sonder, 1999), motivation (Klatter et al., 2001), strategies and styles of learning (Vermunt, 1996; Marton, Hounsell & Entwistle, 1997). This induced to widen the construct of “conceptions of learning”, recently described as a configuration of interrelated beliefs re-garding different aspects of learning. While being described in a wider sense than the original one, the conceptions of learning continue to be associated only with school variables. From a cultural perspective which considers the

* Chile University, Santiago, Chile.** Catholic University of the Sacred Heart, Milano, Italy.*** Catholic University of the Sacred Heart, Brescia, Italy.

Sonia Perèz-Tello et al.

128

role of the general, not only scholastic, environmental conditions, it becomes interesting to highlight possible relationships between such conceptions and extra-school factors, by assuming that the most general “intellectual climate” in which a person lives is not irrelevant to the development of ideas about the goals and the modalities of learning, the elements that facilitate, support it or make it difficult, and so on. Therefore, it is considered important in this study to place the conceptions of learning within a framework that takes into account also the cultural relationships that persons develop out of school.

In fact, the fruition of press publications, television and radio programs, Internet, multimedia software, and so on, puts people in contact with cultural “objects” that produce learning in different, often implicit, ways and with various degrees of intentionality (the prevailing purposes of this kind of frui-tion being pleasure, relaxation, distraction, curiosity or competition).

The scope of the study here described is to relate the conceptions of learning, in terms of multidimensional constructs (constituted by opinions, feelings and attributive operations concerning the outcomes of learning proc-esses), to the experience that students have with a variety of cultural media. To reach such an aim, a wide perspective was adopted, which led to include in the category of “cultural media” tools based on different materials and tech-nologies (from the more traditional papers to the most recent electronic de-vices), of various kinds (information, entertainment, games, communication, advertising, and so forth), and concerning different topics (history, science, art, religion, politics, tourism, fiction, sport, hobbies …). The purpose was to record the frequency of use of television programs, video-cassettes and DVDs, audio-cassettes, discs and compact-disks, radio programs, computer softwares, books, newspapers, magazines, and works sold in instalments as well as the frequency of cinema, theatre and concert attendance and of visits to exhibitions and museums. In a narrow sense, not all these tools can be considered “media”, but nevertheless they are instruments and situations/events through which a person enters into contact (either in a passive attitude of reception or in an active one of application, manipulation, re-elaboration) with cultural issues, with products invented by authors within the canonical codes and designed for public use.

Since these ways of contact vary from one tool to another, also learning – explicit or incidental – that derives will assume various forms. It can there-

Conceptions of learning and use of cultural media

129

fore be hypothesised that a frequent experience and familiarity with particular types of these instruments can be associated with particular beliefs, emotions and attributions regarding the process of learning.

METHOD

MaterialsTwo questionnaires were used. The first one – concerning the concep-

tions of learning – included 49 statements grouped in three sections. The first section regarded the meaning attributed to learning and the conditions that facilitate it (cognitive dimension); the second section referred both to emotional reactions (pleasure, anxiety, irritation, depression, and so on) and to personal commitment (challenge, duty, and so on) associated with school learning (affective dimension); the third section concerned the attribution of both the experiences of failure and success in school activities (attributive dimension). In each part of the questionnaire a series of statements were reported; respondents were asked to rate on a 5-point scale (1 = minimum; 5 = maximum) each statement according to the degree of their agreement. The second questionnaire, regarding the use of media, was composed of 83 items grouped in the following categories: television, video-recorder and/or DVD, computer, radio, audio-cassettes and/or musical compact-disks, books, daily paper and/or magazines, works sold in instalments. A further category con-sidered cinema, theatre, concerts, exhibitions and museums attendance. The answers were asked either in terms of hours of use or in terms of frequency of use estimated on a 5-point scale (from 1 = never to 5 = very often).

ParticipantsBoth questionnaires were administered to 355 students attending sen-

ior high school (II and IV grade). Different types of schools were selected (schools of the arts, technical schools, and professional institutes), distrib-uted in two parts of Italy: North (provinces of Milan and Varese) and center (province of Pesaro and Urbino).


130

Schools were randomly chosen and the students were free to volunteer in the investigation. Girls and boys were distributed in similar percentages with-in each province, within each type of school and at each school level. Table 1 reports the distribution of the participants according to gender and grade.

GRADE

II IV Total

GENDER

boys70 32 102

36.6% 19.5% 28.7%

girls121 132 253

63.4% 80.5% 71.3%

Total 191 164 355

TABLE 1. Distribution of the participants according to gender and grade

RESULTS

Data analysisThe following analyses were carried out. They are described in detail in

distinct paragraphs: • factor analysis of the items of the questionnaire about the conceptions

of learning;• correlations between the factorial scores obtained and the number of

hours of use of the different media and analysis of the variance of the factorial scores according to the number of hours of use of the media;

• correlations between the factorial scores and the frequency of use of each specific kind of media.

Factor analysis of the conceptions of learningThree analyses were conducted, applying the principal component model

with Varimax rotation, to the answers given to the items of each section of the questionnaire about the conceptions of learning. The factors that emerged – coherent with the results from previous researches in which the


131

same questionnaire was employed (Antonietti, Liverta Sempio, Marchetti & Pérez-Tello, 2002a; 2002b) – were the following.

Cognitive dimension The factor analysis carried out on the items corresponding to the first sec-

tion of the questionnaire led to extract five factors that explained 48.02% of the variance.

Factor 1, labelled Relationship with a more experienced model, explained 11.31% of the variance and was loaded by items relevant to learning considered as the observation of actions shown or explained by the teacher. “To show” and “to explain” were conceived as the most important of the teacher`s jobs. The learner, however, was not conceived as an empty container: he/she partici-pates in the learning process with his/her own ideas and knowledge that will successively be submitted to verification.

Factor 2, labelled Cultural production, explained 10.26% of the variance. It is mostly loaded by items that describe learning as a cultural production, emphasis-ing the involvement of feelings. This factor made reference to a kind of learn-ing in which personal ideas are distinguished from the cultural ones, expressed in discussion and social interaction (mainly by means of comparisons).

Factor 3, Individual thought, explained 10.07% of the variance. It was charac-terised by items that intend learning as an intellectual job of individual nature, excluding the possibility to learn through the collaboration with other people.

Factor 4, Reduction of deficits, explained 9.35% of the variance. The items strongly associated to this factor expressed the idea of “lack”; in other words, the student must learn something that he/she is not able to do, and that is why he/she needs someone who teaches him/her how to do it.

Factor 5, Concentration and effort, explained 7.03% of the variance. It was highly loaded by the item “Learning is, above all, matter of concentration and effort”.

Affective dimensionFactor analysis carried out by considering the items of the questionnaire

about the emotional aspects of learning led the extraction of four factors


132

that explained 56.98% of the variance.Factor 1, labelled Negative emotions, explained 20.16% of the variance and

was loaded by items that connect learning to negative feelings such as depres-sion, suffering, hard work, boredom, nervousness, and anxiety.

Factor 2, Opportunity and self-efficacy, explained 17.71% of the variance. It was linked to items that describe positive feelings associated with learning: learning is felt as an opportunity to estimate and to increase one’s own abili-ties. Such an opportunity is handled with much confidence and success by the person him/herself. At the same time learning is felt as a moment of growth and personal change.

Factor 3, Will and pleasure, explained 11.23% of the variance and described learning as a voluntary act accompanied by pleasure, as well as an interesting occasion.

Factor 4, Duty and challenge, explained 7.89% of the variance and it was exclusively composed by the items that expressed learning as a duty, not in a negative way, but as a challenge.

Attributive dimensionThis dimension regards the section of the questionnaire related to the at-

tribution of failure in school situations, to the possibility to learn from the mistakes, to the teachers’ attribution of the students’ errors (according to the vision of the students), and the feeling caused by success in school tests. Factor analysis performed on items of this section yielded five factors that explained 56.04% of the variance.

Factor 1, Success as well-being, explained 12.98% of the variance and only included items concerning success. Such success was related to different as-pects: the teacher, the subject matter, the personal duty, and the classmates.

Factor 2, labelled External attribution of failure according to teachers, explained 11.51% of the variance and referred only to items related to the teacher`s at-tribution of failure. The errors made by the students are always attributed to the teachers in an external way (depending on the assignments given or even on the teachers themselves), but not expressly on the student.

Factor 3, External attribution of success, explained 11.33% of total variance and was loaded by two items related to the success attributed to external as-


133

pects such as chance and surprise.Factor 4, External attribution of failure, explained 10.18% of the variance.

It was composed of items regarding failure in school situations, describing failure as being dependent on external aspects such as the teacher and the characteristics of the task.

Factor 5, Internal attribution of failure and success, explained 10.04% of the variance. It described both failure and success as being dependent on the student and on his/her own abilities.

Conceptions of learning and number of hours of media useFirstly correlation coefficients (Spearman’s rho) between factorial scores

and hours of use of each media were computed. Secondly, analyses of the variance, assuming factorial scores as depending variables and the numbers of hours of use of each media as independent variables, were carried out. To reach such an aim, the distributions of the answers regarding the number of hours were separated into three groups, so differentiating participants ac-cording to a low, medium or high use of each type of media.

In the cognitive dimension, Factor 2 (learning as cultural production) was pos-itively correlated to the hours spent reading books, magazines or daily papers and to the fruition of exhibitions or museums, and was negatively correlated to the number of hours spent on the computer (respectively, rho = .21, p < .01; rho = .15, p < .05; rho = -.13, p < .05). Statistically significant differences occurred only with reference to book reading and exhibition going: this fac-tor is higher in the group of students that read for a greater number of hours (from 3 to 8 hours a day) (F = 4.06, p < .05) and often visit exhibitions and museums (F = 5.76, p < .05).

In the affective dimension, Factor 1 (negative emotions) was more present in the group of students that never go to theatre (F = 4.35, p < .05) and that read less (from 0 to 1 hour a day) (F = 4.13, p < .05); it was negatively corre-lated to the number of hours spent reading books, magazines or daily papers (rho = -.15, p < .01). Factor 2 (opportunity and self-efficacy) was positively correlated to reading (rho = .12, p < .05) and negatively correlated to the number of hours spent in front of the television (rho = -.13, p < .05); it was also higher in students who visit exhibitions and museums frequently (F =


134

6.18, p < .05) and in those who go often to the cinema (F = 6.03, p < .01). Factor 3 (learning as will and pleasure) and Factor 4 (duty and challenge) were also correlated to the number of hours of reading (rho = .14, p < .05). Factor 4 was more present in the group of students that read for many hours a day (from 3 to 8 hours) (F = 4.26, p < .05), whereas Factor 3 was significantly more present in those who use the computer less (F = 3.,24, p < .05) and in those who often go to the cinema (F = 4.18 ,p < .05). Both factors were mostly present in students who frequently visit exhibitions and museums (re-spectively, F = 15.41 and 5.25, p < .01) and go to the theatre (F = 25.28, p < .01). In the attributive dimension, Factor 1 (success as well-being) was positively correlated to the number of daily hours reading books, magazines and daily papers (rho = .19, p < .01). This factor was mostly present in students who read a lot (F = 4.65, p < .01) and often visit museums, exhibitions (F = 8.25, p < .01) and go to the theatre (F = 5.74, p < .01). Factor 2 (external attribu-tion of failure according to teachers) was correlated to the number of hours of computer use (rho = .16, p < .01), and was higher in students who often go to concerts (F = 4.36, p < .05). By contrast, it was negatively correlated to the hours spent reading books, daily papers and magazines (rho = -0.13, p < .05). Factor 5 (internal attribution of both failure and success) was positively correlated to the number of daily hours reading books, magazines and daily papers (rho = .16, p < .01) and was more present in those who read for much time (F = 5.10, p < .01), use the videorecorder less (F = 3.29, p < .05) and seldom listen to the radio (F = 3.59, p < .05).

Conceptions of learning and frequency of use of specific types of mediaTo describe the existing relationships between the conceptions of learn-

ing and the use of specific types of cultural tools, Spearman’s coefficients were computed by considering the factorial scores of each dimension of conceptions of learning and the evaluations of frequency of use of every media type. Such evaluations were then summed up within the same media category; for example, evaluations about all the considered types of television programs (films, cartoons, information news, soap operas, talk shows, adver-tising, and so on) in order to obtain a comprehensive evaluation of frequency


135

of use of that media category. Such overall evaluation was complementary to the estimate of daily hour media use. Correlation coefficients were computed also by considering the factorial scores and these overall evaluation scores.

Cognitive dimensionFactor 1 (learning as relationship with a more experienced model) was cor-

related to the overall evaluation of frequency of video-recorder and DVD use (rho = .18, p < .01), daily paper, magazine and newspaper reading (rho = 0.16, p < .01), television watching (rho = .15, p < .05), collecting and reading works sold in instalments (rho = .13, p < .05). More specifically, this factor was relat-ed to the use of video-cassettes and DVD concerning medicine and science, music and film (respectively, rho = .16 and .15, p < .01; rho = .14, p < .05) and to the reading of daily papers and sporting magazines (rho = .12 and .14, p < .05). This factor was correlated to culture, medicine and science, sport, music, television programs (respectively, rho = .15, .18, and .18, p < .01). The types of works sold in instalments that showed a significant correlation with this factor concerned drama and entertainment shows (rho = .13, p < .05).

Factor 2 (learning as a cultural production) was positively correlated to the overall evaluation of book reading (rho = .22, p < .01) and audio-cassette and compact-disk listening (rho = .13, p < .05), while it was negatively correlated to the frequency of computer use (rho = -.12, p < .05). The specific types of books correlated to this factor were: textbooks, novels and stories, poetry, and foreign language courses (respectively, rho = .21, .20, .18 and .20, p < .01). The audio-cassettes and compact-disks correlated to Factor 2 concerned foreign language courses (rho = .18, p < .01) and audio-books (rho = .15, p < .05). The negative correlation with the use of computer concerned more specifically video-games and leisure, drawing and projecting and mathemati-cal softwares (rho = -.22, -.20, and -.16, p < .01).

Factor 3 (learning as individual thinking) was negatively correlated to the frequency with which television is used for watching films, serials or soap op-eras (rho = -.12 and -.14, p < .05) and was positively correlated to watching cultural, medical and science programs (rho = .14, p < .05).

Factor 4 (learning as a reduction of deficits) was not correlated to any use of media.


136

Factor 5 (learning as concentration and effort) was correlated to the over-all evaluation of frequency of collecting and reading works sold in instal-ments (rho = .17, p < .01), mainly publications about drama and entertain-ment shows , art and tourism, history and culture (rho = .21, p < .01; rho = .13 and .13, p < .05).

Affective dimension Unlike the negative emotions and feeling of duty and challenge associated

with learning (first and last factor of this dimension, that were not correlated to any media), Factor 2 (opportunity and self-efficacy) was correlated to the overall evaluation of frequency of book, work sold in instalments, daily pa-per and magazine reading, as well as to video-cassette, DVD, audio-cassette and compact-disk as well as computer use (rho = .24, .22, .20, .20, .16 and .15, p < .01). Specifically, reading books was correlated to almost all genres (excluding textbooks, novels and stories): historical-cultural books, foreign language courses, practical guides, books on current affairs, poetry, science, religion and spirituality and tourist publications (rho = .21, .14, .14 and .14, p < .01; rho = .13, .12, .12 and .11, p < .05). This factor was correlated to the frequency of use of all kinds of works sold in instalments (except those about art and tourism): publications on science, medicine, health and fitness, history and culture (rho = .18, .15 and .15, p < .01), as well as related to hob-bies leisure and drama (rho = .14, p < .05); moreover, it presented positive correlations to the reading of magazines dedicated to science (rho = .16, p < .01), quizzes, women, current affairs and politics (rho = .13, .13, .11 and .11, p < .05). Factor 2 was associated also with the following types of video-cassette or DVD use: culture, medicine and science (rho = .18, p < .01), music, sport, and foreign languages (rho = .13, .12 and .11, p < .05); the following types of audio-cassettes and compact-disks: audio-books and foreign language cours-es (rho = .18 and .15, p < .01); the following types of computer softwares: e-mail (rho = .16, p < .01) and the Internet in order to find practical and cultural information ( rho = .13 and .11, p < .05). Factor 3 (will and pleasure) was associated with the overall evaluation of the frequency of book reading and audio-cassette and compact-disk listening (rho = .19 and .14, p < .01). All kinds of books, except practical guides (cuisine, sport, collecting, etc.) and


137

tourism publications, were connected with Factor 3: historical-cultural issues, poetry, science, novels and stories, religion and spirituality (rho = .25, .17, .17, .16 and .15, p < .01), and books on current affairs and textbooks (rho = .11 and .11, p < .05). The use of the audio-cassette associated to Factor 3 con-cerned specifically the foreign language courses (rho = .22, p < .01).

Attributive dimensionThe first factor (success as well-being) was correlated to the overall evalu-

ation of the frequency of reading works sold in instalments, books, newspa-pers and magazines (rho = . 26, .22 and .21, p < .01), in watching videotapes or DVDs and listening to audio-cassettes and compact-disks (rho = .16 and .19, p < .01). All kinds of works sold in instalments were significantly corre-lated to Factor 1: science, history and culture, hobbies and leisure, medicine, health and fitness (rho = .23, .22, .20 and .20, p < .01), drama and entertain-ment shows, art and tourism (rho = . 14 and .11, p < .05). All kinds of books, except current affairs and practical guides, were correlated to Factor 1: reli-gion, history, poetry (rho = .23, .17 and .14, p < .01), textbooks , novels and stories, science, foreign languages and tourism (rho = .13, .13, .13, .13 and .11, p < .05). Video-cassettes associated with this factor concerned culture, medicine and science, foreign languages and films (rho = .24 and .16, p < .01; rho = .13, p < .05). Magazines associated with the factor concerned scientific popularization, art and tourism, hobby and leisure, health and fitness (rho = .16 and .15, p < .01; rho = .11 and .11, p < .05). Finally, audio-cassettes asso-ciated to Factor 1 concerned foreign language courses (rho = .22, p < .01).

The second factor (external attribution of failure, according to teachers) was associated with the overall evaluation of the frequency of videorecorder use (rho = .19, p < .01), the reading of works sold in instalments and the use of the computer (rho = . 12 and .12, p < .05). Regarding the videorecord-er, significant associations with the following types emerged: sport, culture, medicine and science, music (rho = .19, .18 and .14, p < .01). Works sold in instalments which showed significant coefficients were: history and culture (rho = .14, p < .01) and drama and entertainment shows (rho = .13, p < .05). Factor 2 was associated to using the computer to construe Web pages, to file or to manage data and to look for cultural information on the Internet (rho


138

= .12, .11 and .11, p < .05). The third factor (external attribution of success) was correlated to the overall estimation of use of the videorecorder and of the DVD (rho = .14, p < .01), specifically to the watching of cartoons, sport and films (rho = .15, p < .01; rho = .14 and .12, p < .05).

Factor 4 (external attribution of failure) was correlated to the reading of magazines about health and fitness (rho = .10, p < .05).

The fifth factor (internal attribution of failure and success) was correlated to the overall evaluation of the use of books, audio-cassettes and compact-disks (rho = .18 and .16, p < .01). The categories of books which turned out to be associated were: novels and stories, textbooks and history (rho = .22, .17 and .16, p < .01). Audio-cassettes significantly associated to Factor 5 con-cerned foreign language courses (rho = .17, p < .01). Tables 2 to 4 summarise the significant effects on factorial scores concerning the various dimensions of the conceptions of learning due to media use and the correlations be-tween factorial scores and frequency of media use found.

FACTOR 1Relationship with a more experienced model

FACTOR 2Cultural production

FACTOR 3Individual thinking

FACTOR 4Reduction of deficit

FACTOR 5Concentration and effort

TV

OVERALLFREQUENCY [c]

SPECIFIC FREQUENCY. culture, medicine and

science [c]. music [c]. sport [c]

SPECIFIC FREQUENCY. films (-)[c]. serial and soap operas (-)[c]. culture, medicine and

science [c]

VIDEO-CASSETTES and DVD

OVERALL FREQUENCY [c]

SPECIFIC FREQUENCY. culture, medicine and

science [c]. music [c]. films [c]

AUDIO-CASSETTES and CDs

OVERAL FREQUENCY [c]

SPECIFICFREQUENCY. foreign language courses [c]. audio-books [c]


139

COMPUTER

NUMBER OF HOURS (-) [c]

OVERAL FREQUENCY (-)[c]

SPECIFIC FREQUENCY. videogames and lei-

sure (-)[c]. drawing and projecting software (-)[c]. mathematical softwa-

re (-)[c]

BOOKS

NUMBER OF HOURS [c] [d]


SPECIFIC FREQUENCY. textbooks [c]. novels and stories

[c]. poetry [c]. foreign language courses [c]

DAILY PAPERS andMAGAZINES


SPECIFIC FREQUENCY

. daily papers [c]

. sporting magazines [c]

WORKS SOLD IN I N S T A L -MENTS


SPECIFIC FREQUENCY. drama and enter-

tainment shows [c]


SPECIFIC FREQUENCY. drama andentertainment shows [c]. art and tourism [c]. history and culture

[c]E X H I B I -TIONS and MUSEUMS

NUMBER OF TIMES [c, d]

(-) negative relationship;[c] significant correlation;[d] significant difference between means (ANOVA).

TABLE 2. Relationships between the cognitive dimension of the conceptions of learning and use of media: Summary of statistically significant results


140

FACTOR 1Negative emotions

FACTOR 2 Opportunity and self-efficacy

FACTOR 3 Will and pleasure

FACTOR 4Duty and challenge

TVNUMBER OF HOURS(-) [c]

VIDEO-CASSETTESand DVD


SPECIFIC FREQUENCY. culture, medicine and sci-

ence [c]. music [c]. sport [c]. foreign language courses [c]

AUDIO-CASSETTESand CDs


SPECIFIC FREQUENCY. audio-books [c]. foreign language courses [c]


SPECIFIC FREQUENCY. foreign language cours-es [c]

COMPUTER


SPECIFIC FREQUENCY. e-mail [c]. Internet (practical info) [c]. Internet (cultural info) [c]

NUMBER OF HOURS (-) [d]

BOOKS

NUMBER OF HOURS(-) [c, d]

NUMBER OF HOURS [c]


SPECIFIC FREQUENCY. history and culture. foreign languages courses. practical guides. current affairs. poetry. science. religion and spirituality. tourism

NUMBER OF HOURS [c]


SPECIFIC FREQUENCY. history and culture. poetry. science. novels and stories. religion and spirituality. current affairs. textbooks

NUMBER OF HOURS [c] [d]

DAILY PAPERS and MAGAZINES


SPECIFIC FREQUENCY. science [c]. quizzes [c]. women [c]. current affairs and politics [c]

WORKS SOLD IN I N S T A L -MENTS


SPECIFIC FREQUENCY. science [c]. medicine, health and fit-

ness [c]. history and culture [c]. hobby and leisure [c]. drama and entertainment shows [c]


141

E X H I B I -TIONS andMUSEUMS

NUMBER OF TIMES [d] NUMBER OF TIMES [d]

NUMBER OF TIMES [d]

CINEMANUMBER OF TIMES [d] NUMBER OF TIMES

(-) [d]

THEATERNUMBER OF TIMES(-) [d]

NUMBER OF TIMES [d]

NUMBER OF TIMES [d]


TABLE 3. Relationships between the affective dimension of the conceptions of learning and use of media: Summary of statistically significant results

FACTOR 1 Success as well-being

FACTOR 2 External attribution of failure, according to teachers

FACTOR 3 Exter-nal attribution of success

FACTOR 4 External attribu-tion of failure

FACTOR 5 Inter-nal attribution of success and failure

VIDEO-CASSETTESand DVD


SPECIFIC REQUENCY. culture, medicine and science [c]. foreign languages courses [c]. films [c]


SPECIFICFREQUENCY. sport [c]. culture, medicine and

science [c]. music [c]


SPECIFICFREQUENCY. cartoons [c]. sport [c]. films [c]

NUMBER OFHOURS (-) [d]

AUDIO-CASSETTESand CDs


SPECIFIC FREQUENCY. foreign language courses [c]


SPECIFIC FREQUENCY. foreign language courses [c]

RADIONUMBER OFHOURS (-) [c, d]

COMPUTER

NUMBER OF HOURS [c, d]


SPECIFIC FREQUENCY. web page construc-

tion [c]. data filling and management [c]. internet (cultural info) [c]


142

BOOKS

N. HOURS [c, d]


SPECIFIC FREQUENCY. religion [c]. history [c]. poetry [c]. textbooks [c]. novels and stories. science [c]. foreign languages courses [c]. tourism [c]

N. HOURS (-) [c] NUMBER OF HOURS [c, d]


SPECIFIC FREQUENCY. novels and stories [c]

. textbooks [c]

. history [c]

DAILY PAPERS andM A G A -ZINES


SPECIFIC FREQUENCY. science [c]. art and tourism [c]. hobby and leisure [c]. health and fitness [c]

SPECIFICFREQUENCY. health and fit-

ness [c]

WORKSSOLDIN I N S T A L -MENTS


SPECIFIC FREQUENCY. science [c]. history and culture [c]

. hobby and leisure [c]

. medicine, health and fitness [c]

. drama and entertain-ment shows [c]

. art and tourism [c]


SPECIFIC FREQUENCY. history and culture

[c]. drama and entertain-

ment shows [c]

E X H I B I -TIONSand MUSE-UMS

NUMBEROF TIMES [d]

THEATERNUMBER OF TIMES [d]

CONCERTSNUMBER OF TIMES [d]


TABLE 4. Relationships between the attributive dimension of the conceptions of learning and use of media: Summary of statistically significant results


143

CONCLUSIONS

The investigation proved the multidimensional nature of the conceptions of learning: the cognitive, affective, and attributive dimensions resulted to be structured in a coherent way, inwardly in a plurality of factors. Moreover, it was shown that each dimension is composed of approximately the same number of factors (from four to five).

In the cognitive dimension the factors can be categorized in factors of a relational and of an individualistic order. The first ones were Factor 1 (rela-tionship with a more expert model), 2 (cultural production), and 4 (reduc-tion of deficit), that foresee the presence of another person in the learning process, a person considered as he/she who presents what must be learned or who interacts with the student in order to induce him/her to learn. The factors of more individualistic order were Factor 3 (individual thinking) and 5 (concentration and effort), that mostly emphasize individual aspects of the learner, his/her gifts and his/her effort.

In the affective dimension two visions of learning emerged: one of a neg-ative order, including disagreeable affects, and the other one of a positive order (learning is felt as an expression of self-affirmation). Factor 1 is a nega-tive factor that gathers emotions and disagreeable states of mind, whereas the others three factors – 2 (opportunity and self-efficacy), 3 (will and pleasure), and 4 (duty and challenge) – regard the feelings of trust in him/herself, and of assertive, pleasant, and personal growth.

The factors of the attributive dimension can be distinguished in two classes: one constituted by factors that recognise the personal involvement in learning, and a second one that instead shows the tendency to elude the truth. Factor 1 – that concerns well-being, happiness, gratefulness, trusting, opening towards the truth and to the others in situations of success – and Factor 5 (internal attribution of failure and success) belong to the first class of factors. The other three factors belong to the second class, which stresses evasion from the truth and the subject`s feeling alien to the educational act, regarded both as a success – as Factor 3 (external attribution of success) indi-cates – or as a failure – just as it is for Factors 2 (thinking that teachers relate student errors to external elements) and 4 (external attribution of failure). In comparison with the picture of a trusting and lively student which lies within

A.Solombrino

143


144

the first class of factors, the picture that emerges from the second class of factors is of someone who does not know him/herself, does not believe him/herself capable of success, so that he/she is surprised and considers him/herself simply as being lucky when he/she succeeds and does not rec-ognize him/herself responsible for the errors that he/she commits.

On the basis of structures that appeared to lie under the conceptions of learning, as revealed by factor analyses, we can predict that relationships be-tween such conceptions and the use of media are not simple to understand. It does not seem possible to correspond a conception of learning to the use of a particular medium. Rather than that, the configurations of various aspects of the conceptions of learning can be associated with the frequent use of a specific medium. For this reason, to draw some conclusions about the relationships between conceptions of learning and use of cultural media, it is useful to have a synoptic picture of the correlations and of the significant differences observed.

In general terms, significant relationships emerged between conceptions of learning and use of cultural media – even though coefficients were not high – in differentiated ways. As far as the cognitive aspects are concerned, Factor 4 (reduction of deficits) was not associated with the use of any me-dia. There were isolated associations for Factors 3 and 5. Instead, there were recurrent associations between use of media and Factors 1 and 2. These as-sociations were mutually excluding: some media (television, video-cassettes, daily papers and sports papers, leisure magazines) were associated with the idea of learning as reception and imitation of an expert model (Factor 1), whereas other media (books and audio-cassettes) with the idea of learning as cultural production (Factor 2). As far as the affective aspects were concerned, it was found that Factor 2 (opportunity and self-efficacy) was correlated to all media except the radio, although sometimes in an opposite direction. We may argue that such kind of emotions constitute a background of the gen-eral experience of employment of cultural instruments, which undeniably offers opportunities to whom makes use of them, generating a perception of competence and contributing to produce a sense of mastery. Besides this common aspect, it appeared that the discriminating aspect is the following: there are media (for example books) whose frequent use was associated with the lack of negative emotions and with a plurality of positive emotions about


145

learning; by contrast, other media (like television) were not accompanied by positive emotions. Finally, as far the attributive aspects were concerned, it was found that the first factor (success as well-being) had a tendency to be shared by several media (except television, radio and computers): the frequent use of such media was associated to success and to positive feelings of competence; this did not happen for media (television and radio) that do not require stu-dents to activate themselves in a particular way (therefore, neither to perceive a particular mastery) unlike the computer that presents a level of difficulty of use which unavoidably leads to experiences of failure. Regarding the other factors, the associations observed with several types of media suggested that the determining element is the distinction between internal and external attri-bution. It appeared that some tools (books, audio-books and audio-cassettes to learn foreign languages) tended to be connected to internal attributions, whereas other tools (video-recorder, radio) to external attributions.

If we consider the media that exemplify various types of culture in a pro-totypical way, it is interesting to observe how the familiarity with the written culture (represented emblematically by books), with the secondary oral cul-ture, of a more passive order (represented by television) and with the culture that is between the writing and the secondary oral culture of a more active order (represented by the computer) are correlated to specific configurations of conceptions of learning.

The reading activity, that implies the exercise of the reflexive thought, was connected with all the three dimensions of the conceptions of learning. In particular, it was positively associated to cognitive, affective, and attributive factors that shape, as a whole, an optimistic picture of the learner. In the opinion of who reads frequently, the learner emerges as an active person, able to produce culture, as someone who grows and expresses him/herself through learning activities, who manages actions of will, who feels pleasure and is comfortable both with him/herself and with others, who is able to observe him/herself critically and to recognize his/her own abilities and his/her own errors. Overall, it deals with a vision of learning that emphasises the subject as an individual, shaping him/her as a person who is “in expansion”.

On the contrary, the use of television – implying higher levels of passivity in comparison to reading – is connected only with factors regarding the cog-nitive dimension of the conceptions of learning. Such factors lead to consid-


146

er the latter type of learner to be more limited than the former. He/she who learns is, in the opinion of he/she who often watches television, a spectator, someone who observes or listens to what the others do or say. Nevertheless, it deals with a person that shows a certain degree of activity, since he/she is capable of evaluating his/her own actions and of recognising his/her own intellectual skills when television watching is oriented towards cultural and scientific programs and not towards fiction. Moreover, this conception of learning is more relational than the previous one, since he/she who learns is seen mainly as engaged in interpersonal relationships, concentrated on hear-ing or observing others. If reading books is connected with the picture of the learner as the protagonist of the learning scene, the use of television is connected with the picture of the learner as someone who shares the learning scene with any other.

The computer should be associated to conceptions of learning according to which the learner should be seen as someone who is part of a distributed activity of knowledge. By contrast, the use of the computer was connected to a few factors that, on the whole, give the picture of the learner as being alien to culture (regarding production and mistake making).

In fact, the use of the computer is negatively correlated only with the cog-nitive factor of learning seen as production of culture, that is with the idea of a person inserted in a historical-cultural community, and it is positively associated with the attributive factor “external attribution of failure accord-ing to teachers”, that brings to see the learner as someone whose errors do not depend on him/her. The few and incoherent connections of the use of the computer with the factors related to the conceptions of learning could be a result of a more reduced use of this media in comparison to books and television, and which therefore does not consent this tool to have an incisive influence.

The investigation opened interesting questions because of the negative correlations found. The factor that describes learning as negative emotions, boredom, suffering and anxiety was not associated with the use of any spe-cific media. The relation between the affective way of living learning and “cultural curiosity” seems to be confirmed by the relationship between the factor that looks positively upon learning, regarded as an occasion of growth based on one’s own trust, and the use of all media (excluding television, that


147

seems to be linked to tendencies of “passivity”) Conceiving learning as an individual intellectual action was not associated

with a flexibility of use of media, since who shared such conceptions tended to attribute importance to watching cultural and scientific programmes only. The individual relationship with learning holds a relationship with culture, but has a passive connotation.

Data evidenced significant relations between conceptions of learning and the use of media, but the question about the direction and the kind of pos-sible influences remains open. Is it the frequent use of one kind of cultural instrument that causes the development of certain conceptions of learning or is it the convictions about learning that lead individuals to prefer certain cultural instruments that are in line with those convictions? As an example: a relationship was found between the reading of works in instalments and emotions of concentration and effort.

Does this relationship derive from the fact that people collecting instal-ments are induced – thanks to the perseverance and sistematicity required in the collecting process, and so on – to believe that, in order to learn, a constant effort is important? Or is this conception of learning that induces people to accept the task of collecting, within an “auto-didactic” perspective? The rela-tionship between conceptions of learning and the use of cultural instruments does not necessarily has to be linear; it could be circular, that is, a relation of reciprocal influence.

For example, the association between use of periodicals and the concep-tion of learning as the relationship with an expert model might be the re-sult of the fact that who has such a conviction about learning tries to keep him/herself informed by reading texts in which competent commentators describe facts; in turn, the habit of reading this kind of texts strengthens the idea that, in order to learn, it is necessary to read what the experts say and think.

The ideas that one has about learning orients people towards particular kinds of use of cultural media. The experience had with certain instruments and cultural objects induces people to develop particular convictions about learning: the conceptions of learning can be influenced by behaviours that are part of the cultural matrix in which the conceptions develop themselves.


148

REFERENCES

Antonietti, A., Liverta Sempio, O., Marchetti, A., Perez-Tello, S. (2002a) Con-cezioni dell’apprendimento dalla scuola media inferiore all’università, XVI Congresso Nazionale della sezione di Psicologia dello Sviluppo dell’Asso-ciazione Italiana di Psicologia (AIP), Bellaria, 15-17 settembre 2002, volu-me dei riassunti: pagine 287-289.

Antonietti, A., Liverta Sempio, O., Marchetti, A., Perez-Tello, S. (2002b) Le concezioni dell’apprendimento degli studenti di scuola superiore, IV Con-gresso Nazionale della Società Italiana di Psicologia dell’Educazione e del-la Formazione (SIPEF), “Gli psicologi nei contesti educativi: ricerca e for-mazione”, Urbino, 5-7 dicembre 2002, volume dei riassunti: pagina 110.

Klatter, E. B., Lodewijks, H. G. L. C., Aarnoutse, C. A. J. (2001) Learning conceptions of young students in the final year of primary education. Learning and Instruction, 11, 485-516.

Lonka, K., Lindblom, Y. (1986) Epistemologies, conceptions of learning and study practices in medicine and psychology. Higher Education, 31, 5-24.

Marton, F., Hounsell, D., Entwistle, N. (1997) The experience of learning. Edin-bourgh: Scottish Academic Press.

Marton, F., Saljo, R. (1976) On qualitative differences in learning. I: Outcome and process. British Journal Educational Psychology, 46, 4-11.

Slaats, A., Loderwijks H., Van der Sonder, J. (1999) Learning styles in second-ary vocational education: Disciplinary differences. Learning and Instruction, 9, 475-492.

Van Rossum, E. J., Schenk, S. M. (1984) The relationship between learning conception, study, strategy and learning outcome. British Journal of Educa-tional Psychology, 54, 73-83.

Vermunt, J. D. H. M. (1996) Metacognitive, cognitive and affective aspects of learning styles and strategies: a phenomenographic analysis. Higher Educa-tion, 31, 25-50.


149

INTELLIGENCE, EDUCATIONAL PRACTICES AND SCHOOL REFORM:

ORGANISATIONS CHANGE, REPRESENTATIONS PERSIST

Felice Carugati*, Patrizia Selleri*1

INTRODUCTION

Since at least the year 2000, the Italian national system of education (in every levels of its organization, including University) has been the object of significant legislative changes, also due to political and governmental change in the country. One of legislative change does concern the introduction of the normative parity between public schools (governed by the public national system i.e. by Ministry of Instruction, University and Research – MIUR) and private schools (i.e. schools which are governed either by Catholic or private organizations). The issue of parity between the two systems is of major in-terest at two levels: one of them is ideological (privatisation of education, confessional vs. laic education); the second is that the law allows private or-ganisations to be funded either directly or indirectly, by the State and by local authorities, due to the public interest for education.

So, it follows that a period of political and organizational changes could be regarded as a social laboratory where a rich amount of social phenomena are subject of scientific scrutiny by using adequate conceptual and empirical tools.

As we have already mentioned, the law acknowledges all actors a equal dignity in participating at the building of the educational system, but it is well known that in Italy the debate between laical and religious actors about the main aims of education, the centrality of personhood, the family and the

* Department of Education – Faculty of Psychology, Alma Mater Studiorum Università di Bologna, Italy. 1 This research was supported by a grant from Italian Ministry of Instruction, University and Research (MIUR: ex- 40% 2000-2002).

Felice Carugati, Patrizia Selleri

150

child, the best tools both for teaching and for educating children to the citi-zenship. Moreover our studies on social representations of intelligence and education (Mugny & Carugati, 1985) have illustrated to what extent teachers are a special category of people questioned by the relative inexplicability of phenomena of intelligence and education.

A brief overview of this research is useful for introducing the empirical section of our contrubition.

Our theoretical starting point is the theory of social representations, and particularly the specific empirical approach, which has been developing since 20 years by the authors (Mugny & Carugati,1985, 1988, 1989; Carugati, 1990; Carugati & Selleri, 2001; Carugati, Selleri & Scappini,1994).

First, what actually are social representations? Summing up a vast amount of research which began with the Serge Moscovici’s masterpiece about the psychoanalysis in French culture (1960) it could be said that every repre-sentation tends to turn or transform an unfamiliar thing (for instance the scientific object like psychoanalysis) or the unfamiliar in general into some-thing familiar. Consensual universes are universes where each of us wants to feel at home, sheltered from areas o disagreement and form incompatibility. Whatever is said or done tends to reconfirm the accepted assumptions and meanings, to affirm rather than contradict. Novelty as such is accepted to the extent that it maintains some liveliness. We are confronted with the dynamics of familiarization of the strange whereby objects, individuals, and events are recognized and understood on the basis of prior encounters or models. As a result, memory, tends to predominate over logic, the past over the present, the verdict over the trial. Our reliance on the familiar as the preferred refer-ence point is an equally widespread phenomenon. It serves as our standard of comparison for everything that happens and is observed. We not only learn what characterize a ‘decent person’, a ‘smart student’, a ‘brilliant researcher’ but in addition this shared knowledge is useful as a criterion for judging the unfamiliar, the abnormal, etc.

The basic tension between the familiar and the unfamiliar is resolved in our everyday consensual universe in favour of the familiar. This is why con-clusions have primacy over premises and the verdict more than the trial domi-nates our social relations. Before seeing and listening to someone, we make a judgment about him/her, we categorize him/her and we form a mental im-age of him/her. Such mental categories are not merely cognitive abstractions,

Intelligence, educational practices and school reform

151

but are essentially social in character. By the information we gather and the effort we make in obtaining it, we simply go about confirming this picture. Johnson-Laird and Wason, 30 years ago, supported this observation:

Subjects seem to be biased towards attempting to verify a conclusion, whether it is their own initial answer or one given to them by experimenter to evaluate. They seek to determine whether the premises could be combined in such a way as to render the conclusions true. Of course, this merely shows that conclusions and premises are consistent, not that the conclusion follows from the premises (1977, 157).

The sequence of arguments or deductions commonly presented in labo-ratories, as in daily life, seems to serve as a way of proving things; the truth is that it merely repeats the pre-existing path (or bias) which is known from the start. The representations we shape, for instance about a salient topic like intelligence always reflect an incessant effort to turn something unfamiliar to us into something ordinary and immediately present.

As a result of successive small alterations what was once remote now seems close to us and what was once experienced as abstract is now concrete and almost an everyday thing.

Now it is time to illustrate an example of these general considerations about the interplay of science and common sense, taking as a case in point a issue which seem to us is underlying every reform in education. Let us begin with some questions concerning the issues of learning, teaching, instruction: why some students don’t learn or at least don’t learn according to research-ers’ or teachers’ expectations? How these expectations are produced? It may be supposed that a huge amount of the tremendous work of previous and recent research is inspired by the aim of giving some answers to these ques-tions, i.e. of coping with the challenge of improving learning and fostering the will to learn.

From the point of view of both researchers and teachers (be the teachers formal or informal ones) and from the point of view of a curious and inter-ested observer, improving learning and fostering the will to learn is a suitable endeavour only under the condition that the learners are attributed some faculty or ability to learn. In other words, the learner is assumed (at any point of his/her life course) either by scholars or teachers, to be ready to access some specific content of instruction through some specific procedures (i.e. the procedures for learning).


152

We are confronted with two possible outcomes: the learner does meet the expectations and he/she is successful or not. In both cases where success or failure come from? In other words: is learning and its counterpart (failure) the phenotype of any genotype?

Do the two phenotypes come from a unique (common) genotype? In both cases (but particularly in case of failure) scholars and/or teachers are obliged to explain/ justify the success or the failure: some kind of explanation/justi-fication should be formulated.

We use the metaphor of genotype because we assume that in the history of psychology and particularly of developmental and educational psychol-ogy, a possible common root of success/failure and of the triangulation be-tween instruction/teaching/learning is the phenomenon of intelligence, with unceasingly debates about the nature/culture origin of intelligence.

In this line, the shift of views about intelligence parallels the mainstream shift in metaphors of mind from natural sciences to computer sciences. Moreover, according to Sigel (1986) psychologists seem nowadays to wallow increasingly in diversity and fragmentation not only of subject matter but also in terms of the way the intelligence is studied.

There is thus something of a paradox in so far as experts both agree and disagree about what intelligence is on classical, dilemmatic and across dec-ades dimensions: the one versus manifold; the nature versus culture.

One explanation may be that experts have accentuated the diversity and fragmentation of their work, as a consequence of some kind of logic of de-velopment of scientific endeavour. A complementary explanation may be that experts tend to disagree among themselves in the course of inventing theo-ries (or even must disagree, in order to produce and maintain professional dis-tinctiveness!), while converging in their views when called upon to presents themselves to colleagues in symposia or fair-play-like situations or when they have to diffuse their theories to lay persons or amateurs. A third explanation would be that science is not a unitary discourse, but reflecting different gen-eral Weltanschaungen.

The investigation of intelligence as an educational phenomenon has dif-ferent aims and objectives from the investigation of intelligence as a problem for cognitive science. An example of the effort to match the aims of educa-tion and of cognitive science perspective is a pleading in favour of a theory of ‘multiple intelligences’ (Gardner, 1983) that spans the range of previous


153

theories and the claims of nature – culture dilemma. For instance, Kornhab-er, Krechevsky and Gardner (1990) assumed that the key features of almost all definitions of intelligence have no less to do with the competences inside one’s head than with the values and opportunities afforded by society to en-gage these competences. Even in a universally developing competence like language, it is only thanks to the interaction between adult and child that such a faculty develops. Societies teach their children those bodies of facts, theories, skills, and methods that comprise its various domains of knowledge, ranging from physics to fishing.

Discussing the different views held in Japan vs. USA cultures, the authors underline that in Japan the development of intelligence is fostered by widely shared values (i.e. school, achievement and diligent study), which in turn are supported by the institutions of the Japanese society. Furthermore, societal values support both schooling and an emphasis on effort and intrinsic moti-vation rather than on innate abilities.

This positive picture of Japan is counterbalanced with respect to USA. America’s break with tradition-bound learning coincident with its love for new technologies may have rendered it especially vulnerable to explanations of intelligence borrowed by hereditary views and the eugenics movement to take hold in USA. Science with its paper-and-pencil tests, and factor analy-sis supported views that White, Anglo-Saxon, Protestant people possess the most intelligence. In the minds of many, these people came from the best genetic stock. Thus America came to believe that intelligence was born and not made. As Stephen Gould (1981, 147) has already noted, the rendering of the concept of intelligence into a reified, inherited trait was an ‘American invention’. This careful, critical diagnosis of the ways intelligence is a matter of interplay of cultural shared values and the historical vicissitudes of these values is a good example of the central value-laden role intelligence has in modern societies, apart from the effort scholars do in fact to describe and explain it in terms of academic language.

If this is a sketch of the academic views about intelligence, we can see now which is its counterpart in the modern commons sense universe of some people who are particularly interested in the topic: parents and teachers, and particularly mothers and teachers who are at the same time mothers.

Why these specific categories of women? Our first concern is with the nature of the information available about the actual object of representation.


154

The type of information may be formal or informal. Mothers for instance (but fathers too, and perhaps more than their wives or partners), might not be experts about intelligence; their main sources may be personal experi-ence (the everyday experience of being mothers, media exposure, reading of magazines, etc.). The knowledge of mothers consists of unsystematically collected information obtained through `day to day’ participation at the dis-courses, which circulate in their culture.

Some other people, on the other hand, acquire information through a more systematic source. Teachers, for example, in the course of their pro-fessional education, have access to organised information, such as psycho-logical and sociological models of development, teaching and learning. Thus mothers may be seen to fall into a category between laypersons and teachers. They have specific reasons for constructing and sharing representations of intelligence, and teaching and learning methods. Both parents and teachers have also to cope with the question of interindividual differences about intel-ligence, the origins of these differences, and the question about the success and failure at school.

The question of interindividual differences concerning intelligence and their origins is a very old one and one which, over the centuries has intrigued philosophers, psychologists and every kind of maîtres à penser, giving rise to endless controversies. Intelligence emerges as one of the most predominant positive values of Western society; no wonder, therefore, that a polysemantic discourse, both by lay people and scientists is continuously being built and rebuilt. But while scientists could afford the luxury of keeping themselves engaged in endless debates on this topic, parents and teachers have neither the freedom nor the time for such contemplation, nor do they have the time to wait for the ultimate scientific explanation. Teachers and parents are very concerned by dealing with everyday problems relating to school and the huge amount of responsibilities at home that they encounter when children and pupils are involved.

Like people confronted with the unfamiliarity and the relative inexplica-bility of the psychoanalysis, and illness (as in the classical studies on social representations: Moscovici, 1961; Herzlich, 1972; Jodelet, 1989) parents and teachers confronted with interindividual differences, may feel a gap between the information at their disposal and the information necessary to give meaning and justification for this phenomenon.


155

In more theoretical terms, parents and teachers as social categories are very sensitive to the relevance of questions such as the origins of interindividual differences in intelligence and school performances.

While they may lack a complete grasp of the subject, the concrete and eve-ryday problems and necessities, which they encounter about intelligence and school performances, force them into making inferences and decisions. And, indeed, it is the coupling of a relative lack of information with the pressure to make decisions, which is a principal characteristic of the socio-cognitive activities of these social categories.

Thus the first set of hypotheses we put forward is inspired by Moscovici’s suggestion that social representations tend to turn an unfamiliar thing (or the unfamiliar in general) into something familiar. For testing our hypothesis, we have an object: the existence of interindividual differences in intelligence, and the inexplicability (unfamiliarity) may be explored by the degree of weakness attributed to the scientific explanations. A complementary way of exploring the effects of unfamiliarity may be found in a lack of competing explanatory models held by subjects.

The second set of hypotheses is devoted to the relationships between social representations and social identity: more specifically, the experience of conflicts of identification is conceived as a factor influencing some transfor-mations in social representations. Concretely speaking, teachers may also be parents. As teachers, they are compelled to try and prevent pupils failing in school; as parents, they are protective towards their children when necessary in school matters. So, in a way, they are protecting their children against them-selves. A second example is the professional working mother.

As a housewife, a mother may tend to explain intelligence and school performance as a product of her direct involvement in child rearing and as a task socially assigned to mothers. However, as a working mother, she may feel guilty about not being involved enough in the up bringing of her children. In both cases, that of the parent who is a teacher and the working mother, spe-cific socio-professional positions may induce identity conflicts, which may be resolved by re-adjusting their representations of intelligence. We do not, by any means, conceive of these re- adjustments as biases or errors as the social cognition approach suggests.

On the contrary, the hypotheses we have sketched suggest that the func-tions and the transformations of social representations are located at the


156

interface between the individual and society. Transformations of social rep-resentations occur when the unfamiliarity and the inexplicability of interin-dividual differences become salient for particular social categories and when this topic activates a conflict of identification.

The hypothesis that social representations theory is an approach that might account for how people cope with the dynamics between what they don’t know, what they already know and what they should know for a ‘prac-tical’ decision making, particularly when they have to make decisions about important issues, was the main hypothesis for the first empirical investigation in the domain of intelligence and its development (Mugny & Carugati, 1985, 1989).

As far as people feel interindividual differences in intelligence and de-velopment both as unfamiliar issue and an inexplicable question (and they are not confident in scientific explanations), they produce a sophisticated discourse about interindividual differences, borrowing contents from both ‘ethnotheories’ and academic theories (Carugati & Selleri, 1995): it concerns intellectual abilities and performances as a gift (as a gift and not as a matter of development!); intelligence is viewed as a matter of both logical abilities and conformity to everyday social rules; educational practices as a matter of pressure on the pupil and of strict assessment at school; moreover, explicit attribution of responsibility to teachers for the development of intelligence and for school failure is put forward. This figure is even more evident among specific categories of people: parents, teachers, and teachers who are parents at the same time.

A further result may be summarized as follows: intelligence is by no means a monolithic notion, whose meaning originates from the experts’ work and then inoculated into the lay people’s discourse: on the contrary, the cogni-tive monophasia of different theories, that academic scholars build, defend and hold (Carugati, 1990a, 1990b; Carugati & Selleri, 1995) is the raw material used by lay people as building blocks for a multiplicity of different discourses and for a diversity of socio-cognitive functions.

People manifest systematic patterns of discourse as responses to the spe-cific social positions they hold in social systems and with reference to signifi-cant turning points in their lives.

Thus the cognitive monophasia of experts shifts to a well organized cogni-


157

tive polyphasia in lay people’s discourses, which serve for them to negotiate between competing theories according to their positions in the social system and to their relations to the inexplicable topic: the case of parents-teachers who defend (more than non parents- non teachers) a conception of intel-ligence as a gift unequally distributed among children is a nice example of how people construe their cognitions not in a vacuum but in the social arena, where conflicting social identities (parents, teachers) are questioned by a sub-jectively inexplicable topic.

AIMS OF THE STUDY

According to our general framework, the aim of this contribution is to verify 20 years later and with a wider sample of teachers, the hypothesis of the influence of inexplicability of intelligence as anchoring point in building up a representation if it in terms of a gift unequally distributed among pu-pils. A complementary aim is to illustrate to what extent teachers working in public vs. private schools do differ in sharing an interpretation of intelligence in terms of gift.

In other terms, we try to confirm our previous results with a sample of teachers working in elementary, junior high school and high school (in previ-ous research only elementary school teachers were studied). Furthermore as we have already shown, representations of intelligence are related to educa-tional practices supposed more or less effective in coping with difficulties in learning. We remember here that the main result of our previous research was that teachers who share a conception of intelligence as inexplicable one, do organise their conceptions of intelligence in terms of gift and they are more confident in educational practices in terms of severe evaluation and competition.

What about teachers with the above mentioned social characteristics? Is the inexplicability a consistent socio-cognitive anchoring point? Our expec-tation is yes, while we suppose that no specific differences would exist for teachers working in private vs. public schools. Eventually it is possible that private teachers would underline homework and competition as a issue of the liberal conception, which emphasises personal commitment and individual responsibility in learning.


158

METHOD

This study is a part of a wider research with an Italian sample of teachers (N = 1,232; the sample is a non-probabilistic sample, which allows to com-pare systematically public vs. private teachers); a short version of our ques-tionnaire on intelligence and educational practices (Mugny & Carugati 1985) has been submitted to the sample. Because a obvious variable does concern differences between men and women and between men and women teachers, we will limit here to present results concerning a sub-sample of 952 women teachers working in three levels of schools: primary, middle, and high school and in private and public type of schools.

Primary school Middle school High school Total

Type Public school 192 150 168 510

Private school 174 130 138 442

Total 366 280 306 952

TABLE 1. Italian sample of women teachers

The items of the questionnaire we used for statistical treatment are: intel-ligence (Q1: 11 items, five point scale); educational practices (Q2: 15 items, five point scale). They are a sample of items we used in previous research (Mugny & Carugati, 1985), sample, which has been shown as the most repre-sentative of the organisation of teachers’ representations (cf. Selleri, Carugati & Scappini, 1995).

RESULTS

Through two separate factor analyses (Q1; Q2) we have verified the con-sistency of a theory of intelligence as a gift and natural inequalities (Q1, F 1); a complementary conception on intelligence as matter of logic and revealed by school success (Q1, F 3), while the Q1, F2 is consistent with a complex conception where intelligent pupils come from families whose parents value intelligence, intelligence is an instrument to conform to societal norms, and assessing school work is the best way to promote intelligence.


159

Q1F1: Gift and natural inequalities (% of variance explained before rotation: 21,841; after rotation 11,700; items are presented with loading >.400)

Intelligence doesn’t develop: It is a hereditary gift .770Some people are born with more intelligence, others with less .473

Q1F2: Family, conformism, evaluation (% of variance explained before rotation: 5,890; after rotation: 11,263; ; items are presented with loading >.400))

Intelligent children come from families where parents value intelligence .523Being intelligent means conforming to social norms .499The best stimulus to development is school assessment .470

Q1F3: Logic and school success (% of variance explained before rotation: 4,535; after rotation: 9,303; items are presented with loading >.400) )

You have to be intelligent to do well at school .621Logic and mathematics are prototypes of intelligence .580

TABLE 2. Intelligence: Definitions and development (N. Teachers = 952) (FACTOR: Method: Maximum Likelihood; Rotation: Varimax)

As for the factor analysis of Q2, three main factors are shown: F1: improv-ing self-confidence in pupils and psychological atmosphere of classroom, plus improving parents’ responsibility; F2: let pupils work in small groups and with a more expert peer, plus adopting a severe evaluation; F3: giving home work and improving competition between peers.

Q2F1: Confidence, parents, psychological atmosphere (% of variance explained before rotation: 26,555; after rotation: 14,285; items are presented with loading >.400) )

Help the child regain self-confidence .784 Talk to parents about child’s difficulties .588 Check that child really understands the data of a problem .542 Improve the classroom atmosphere .484

Q2Q2F2: Group work and severe evaluation (% of variance explained before rotation : 6,908; after rotation: 13,247; items are presented with loading >.400)

Make the child work in small groups .719Make the child work with a child who is more advanced .584Teach the child to be rigorous in his/her work .479

Q2F3: Homework and competition (% of variance explained before rotation: 2,919; after rotation: 8,851; ; items are presented with loading >.400) )

Make the child compete with other children .671Promise child a reward if he/she does better .499Show him/her that he/she is falling behind other pupils .471Give the child homework .458

TABLE 3. Educational practices (N. Teachers = 952)


160

(FACTOR: Method: Maximum Likelihood; Rotation: Varimax )As a way operationalise the influence of inexplicability of intelligence, we

used the following item: The existence of differences of intelligence between individuals is a mysterious problem, which science is still unable to solve

According to the frequencies of the above mentioned item, a new variable was produced: (MYSTERY) with two modalities: NO MYSTERY i. e. teachers (1-2 frequencies) who doesn’t agree with the content of the item and YES MYSTERY (4-5 frequencies) i.e. teachers who agree that intelligence is inexplicable.

This new variable has been used as independent variable for analysing the influence it has on factors on intelligence and educational practices.

As a consequence this, analysis could be run with a sub sample of teachers (we have excluded both ‘frequency 3 ‘ and ‘missing values’ (this sub-sample is N = 473).

ANOVA F: 10,043; df 3,1; p< .000; Mystery (a) 29,640. p. <.000; type of school (b) .462; p < .497; a * b : .024 p< .877.

FIGURE 1. Intelligence: F 1 Gift and natural inequalities(mystery by type of school: private vs. public school teachers)

3,0

2,8

2,6

2,4

2,2

2,0Public

Mystery

NOYES

Private


161

ANOVA F: 3,551; df 3,1; p < .014; Mystery (a) 8,269; p. <.004; type of school (b) 1,574; p < 0.210; a * b : .170; p< .680.

FIGURE 2. Intelligence: F 2 Severe evaluation (mystery by type of school: private vs. public school teachers)

ANOVA F: 8,378; df 3,1; p < .000; Mystery (a) 15, 548; p <.000;type of school (b) 6,705; p < 010; a * b: 3,874; p< .050.

FIGURE 3. Intelligence: F 3 Logic and school success (mystery by type of school: private vs. public school teachers)

3,2

2,8

3,0

2,6

2,4

2,2Public Private

Mystery

NOYES

2,2

2,0

2,1

1,9

1,8

1,7Public Private

Mystery

NOYES


162

ANOVA F: 1,969; df 3,1; p < .031; Mystery (a) F: ,508; p < .009; type of school (b) F: 1,166; p < .281; a * b: . 307; p < .580.

FIGURE 4. Educational practices: F1 Confidence, parents, psychological atmosphere (mystery by type of school: private vs. public school teachers)

ANOVA: F: .767; df 3,1; p< .513; Mystery (a): F: .304; p< .582; type of school (b): F: .303; p< .582; a * b: F: 1,312; p< .253.

FIGURE 5. Educational practices: F2 Group work, severe evaluation(mystery by type of school: private vs. public school teachers)

4,7

4,5

4,6

4,4

4,3

4,2

Public Private

Mystery

NO

YES

4,0

3,8

3,9

3,7

Public Private

Mystery

NO

YES


163

ANOVA F: 2,979; df 3,1; p < .031; Mystery (a): F: 6,820; p < .009; type of school (b): F: 1,166; p < .281; a * b: F: .307; p < .580.

FIGURE 6. Educational practices: F3 Extra-home work and competition(mystery by type of school: private vs. public school teachers)

DISCUSSION

Educational reforms are open to endless debates, but a core representa-tion of intelligence and educational practices still persists: as for intelligence a trio of anchoring points goes in terms of gift; conformism and evaluation; logic and school success.

As for educational practices: a trio of anchoring points goes around the quality of emotional climate at school and in family; the group work and se-vere evaluation; the extra school work and competition.

The trio, which characterises intelligence and its (possible) development is systematically multimeaning: the theory of gift is complemented by the natural inequalities between people (people are born with differences in intelligence). Moreover family contexts are acknowledged as playing a role in hosting in-telligence, and this role is coupled with conformism and school assessment. Finally, another couple of contents, logic and school success, are invoked.

2,3

2,1

2,2

2,0

1,9

1,8

Public Private

Mystery

NO

YES

1,7

1,6

1,5


164

As for the educational practice the same figure is apparent: family and school socioemtional positive athmosphere; the issue of child’s work once more at school and in family is given a specific meaning by another kind so-cio-emotional athmosphere: the risk (for mediocre pupils) or the benefit (for brilliant pupils) of being evaluated and being put in competition. It could be said the allmost all theories of intelligence and teaching learning methods are well organised by teachers. But what about the relations between these issues and our hypotheses?

The first apparent result is the pervasive influence of the subjective sense of inexplicability of intelligence on the way teachers are positioned on the content of factors. As for the intelligence, the mystery teachers are more prone to the gift, conformism, assessment, and logic and school success. No differ-ence whether they are working in private vs. public schools. A warning should be put forward: some readers would react, saying that the group means are pretty low, particularly for F2 (family, conformism, assessment). Yes we agree, but it is still apparent the difference between the two categories of teachers. It is worth noting that the only interaction effect in the ANOVA is for F3 (Logic and school success) where private mystery teachers are less confident in this is-sue.

The general pattern of influence of inexplicability is shown as for the educational practices but with some modulated evidence: in fact a ‘ceiling ef-fect’ of shared position concerning socio-emotional climate is predominant (cf. F1 group means; but mystery teachers are less confident in the efficacy of that climate for coping with difficulties in learning). No difference as for the second topic (F2: group work and severe evaluation) and a mild effect on the issue of extra homework and competition (F3), where group means are the lowest ones. In this case we observe nevertheless a main effect for both vari-ables (inexplicability and type of school).

CONCLUDING REMARKS

Our first reaction, after analysing these results, was ‘it is too good for being true’. In fact, the overall pattern of 20 year-old results have been confirmed. The socio-cognitive organization of teachers’ representations fits almost per-fectly in with previous results (Mugny & Carugati, 1985). If detailed pat-


165

terns of results concerning the eventual differences between teachers work-ing in three levels of schools (Primary, Middle, High school teachers) will be presented in the next future elsewhere, we underline here that teachers we studied hold well organised and consistent patterns about what intelligence is supposed to be and which educational practices are supposed to be more effective for coping with difficulties in learning.

In a period of institutional reforms it is quite interesting notice that teach-ers working in different socio-institutional (school levels) and ideological contexts (pubic vs. private schools) do share common patterns of represen-tations, which seems to serve as socio-cognitive anchoring points persistent over diversity of work experiences and over years. These results should be integrated with other fine grained analyses, but they still offer a twofold line of considerations.

First, at least in the case of teachers, representations they share about in-telligence and educational practices may be a common referent point when they are seen as explicitly diverging or even conflicting at the ideological level about the ultimate aims of school and education. The political and trade un-ions debate is a case in point.

Second, at a theoretical level, social representations could be interpreted as a interface between everyday experiences and the ideologies and wider Weltanschaungen, which play an important role in the dynamics of construction of symbolic consensual universe in modern cultures.

REFERENCES

Carugati, F. (1996) Intelligence, school, marks, and pupils: A rejoinder to Räty and Snellman’s (1995) ‘A social fabric of intelligence’, “Papers on Social Representations”, 5, 2, 151-155.

Carugati, F., & Selleri, P. (1998) Social representations and development: Ex-perts’ and parents’ discourses about a puzzling issue. In U. Flick (Ed.) The psychology of the social. Cambridge, Cambridge University Press, pp. 170-185.

Carugati, F. (1990a) Everyday ideas, theoretical models, and social represen-tations: The case of intelligence and its development. In G.R.Semin and K.J.Gergen (Eds.) Everyday understanding: Social and scientific implica-


166

tions. London: Sage, pp. 130-150.Carugati, F. (1990b) From social cognition to social representations in the

study of intelligence. In G.M.Duveen, B. Lloyd (Eds.) Social representa-tions and the development of knowledge. Cambridge: Cambridge Univer-sity Press, pp. 126-143.

Carugati, F. (1992) L’intelligenza fra scienza e senso comune. In W. Fornasa, A. Vanni (Eds.) Con voce di nido, Milano, Franco Angeli., pp . 34-53.

Carugati, F., Emiliani, F. & Molinari,L. (1990) Being mothers is not enough: Theories and images in the social representations of childhood. “Revue Internationale de Psychologie Sociale”, 3, 287-304

Carugati, F., Emiliani, F., Molinari, L. (1992) De l’enfant en général à l’enfant propre: jeux d’images et conceptions du développement. “Bulletin de Psy-chologie”, XLV, 405, 232-238.

Carugati, F. & Nizzoli, U. (1988) Il fili della crisalide, Abano Terme, Francisci Editore, pp. 228.

Carugati, F. & Selleri, P. (1992) L’intelligenza al plurale: un punto di accordo fra scienza, educazione e senso comune?. In L. Grossele (Ed.) Un piano di studi libero per la scuola elementare, Padova, Centro Studi ‘La Chiave’, pp. 3-15.

Carugati, F. & Selleri, P. (1995). Diskurse von Eltern und Experten über En-twicklung. In U. Flick (Ed.) Psychologie des Sozialen, Reinbeck, Rowohlts Enziklopädie., pp. 218-239.

Carugati, F. & Selleri, P. (2000) Pratiques éducatives, socialisation et représen-tations sociales, In C. Garnier, M.L. Rouquette (Eds.) Représentations so-ciales et éducation. Montréal, Editions Nouvelles, pp. 1-25.

Carugati, F., Selleri, P. & Scappini, E. (1994) Are social representations an architecture of cognitions? A tentative model for extending the dialogue, “Papers on Social Representations”, 3,2, 134- 151.

Emiliani, F., Lorenzi-Cioldi, F. & Carugati,F. (1989) Modèles de comporte-ment et images de la crèche chez des éducatrices de jeunes enfants. “Cah-iers de Psychologie Cognitive”, 1986, 6, (6), 633-655.

Fernandes Prata, A., Carugati, F. & Selleri, P. (2001) Sobre a avaliação: O comportamento subjacente à ‘norma de referẽncia, “Anàlise Psicològica”, XIX, 3, 435-452

Gardner, H. (1983) Frames of mind: The theory of multiple intelligences. New Yourk: Basic Books.


167

Gould, S. (1981) The mismeasure of man. New York: Norton. Herzlich, C. (1972). Healt and illness: a social psychology analysis. London:

Academic Press.Jodelet, D. (1989). Folie et représentations sociales. Paris: PUF. (Eng. trans.

(1991) Madness and social representations. Hempel Hampstead: Harvest-er- Wheatsheaf

Johnson-Laird, P.N. & P.C. Wason P.C. (eds.) (1977) Thinking : readings in cognitive science. Cambridge: Cambridge University Press.

Kessen, W. (1979) The American child and other cultural inventions. Ameri-can Psychologist, 15, 815-820

Kornhaber, M., Krechewsky, M. & Gardner,H. (1990) Engaging intelligence. “Educational Psychologist” , 25 (3 & 4), 177-199

Molinari, L., Emiliani, F. & Carugati, F. (1992) Development according to mothers: A case of social representations. In M. von Cranach, W. Doise, G. Mugny (Eds.) Social representations and the social bases of knowledge. Levingston, N.Y., Toronto, Bern, Göttingen: Hogrefe & Huber Publish-ers, pp. 104-111.

Moscovici, S. (1961) La psychanalyse, son image et son public. Paris: PUF (1976 2.ed.)

Moscovici, S. (1990) Social psychology and developmental psychology: ex-tending the conversation. In G. Duveen & B. Lloyd (Eds.) Social repre-sentations and the development of knowledge. Cambridge: Cambridge University Press.

Mugny, G. & Carugati, F. (1988) L’intelligenza al plurale. Bologna: Coopera-tiva Libraria Universitaria Bolognese., pp. 230.

Mugny, G. & Carugati,F. (1985) L’intelligence au pluriel. Cousset, (Fribourg, Suisse): DelVal, pp .228.

Mugny, G. & Carugati, F. (1989) Social representations of intelligence. Cam-bridge: Cambridge University Press, pp. 196.

Selleri, P., Carugati, F. & Scappini, E. (1995) What marks should I give? A model of the organization of teachers’ judgements of their pupils, “Eu-ropean Journal of Psychology of Education”, X, 1, 25-40.

Selleri, P., Carugati, F. & Bison,I. (1994) Compagni intelligenti e compagni bravi a scuola, “Rassegna di Psicologia”, XI, 2, 29-52.


169

AT A CROSSROADS?THE DEVELOPMENT OF EPISTEMOLOGICAL

BELIEFS AND HISTORICAL THINKING

Liliana Maggioni*, Patricia Alexander*, Bruce VanSledright*

The purpose of this study was to explore changes in teachers’ epistemological beliefs as they participated in a professional development program intended on ex-panding their perspective on historical learning and teaching. Specifically, we devel-oped a questionnaire to evaluate teachers’ beliefs about the learning and teaching of history and examined the factor structure suggested by the data in light of the literature on epistemological beliefs and historical thinking. Then, based on responses on the beliefs measure, we built epistemological profiles and categorized teachers as criterialists, relativists, naïve realists, or dichotomous reasoners. Finally, we monitored teachers’ epistemological shifts following their movements across profiles during the intervention. To frame the discussion, we survey the literature in epistemological beliefs and historical thinking.

“Consider your sowing: you were not made to livelike brutes, but to follow virtue and knowledge”

(Dante Alighieri)

INTRODUCTION

At the climax of Western, medieval culture, the Italian poet Dante imag-ined the speech that prompted Ulysses’s men to cross the threshold of the unknown set by the Pillars of Hercules and to defy the limits of human reason. The history that followed that quest might be read as the similarly passionate, unceasing attempt to overcome the ever expanding boundaries

* University of Maryland.

Liliana Maggioni, Patricia Alexander, Bruce VanSledright

170

of human knowledge, probing the depth of the infinitely small and gazing into the vastness of the infinitely great. A similar quest traversed centuries of Western philosophical thought, investigating the very nature and source of that ever-increasing human knowledge. In this second search, however, modern men and women find themselves less certain than their medieval forefathers, and the echo of Descartes’s doubt still resounds in the debates that cross disciplinary fields as diverse as psychology and history.

If the desire for knowledge still animates human nature, the definition of the relation between knowledge and reality has become much more problem-atic for modern people. The cognitive revolution (Phillips, 1995) emphasized the constructed nature of knowledge, questioning, in its most radical expres-sions, the very existence of a knowable reality external to the individual, and thus theorizing the exclusively subjective nature of knowledge (von Glaser-sfeld, 1991). The issue of objectivity has also been frequently discussed in history, prompting Peter Novick (1988) to provocatively title the introduction to his review of American historiography over the last century, “Nailing jelly to the wall.” In education, and particularly in history education, the challenge to the existence of an objective body of historical knowledge has fueled an already hot debate about curricula content and made the issues regarding the purposes of formal education in the American multicultural society even more complex (Seixas 2000).

Within this field, a line of research broke the dead-lock of the debate about whose history, by dedicating itself to the understanding of what it means to think historically in the first place and what may be the challenges of trans-lating these insights in classroom interventions (Stearns, Seixas & Wineburg, 2000). In so doing, the mainly qualitative research on historical thinking has contributed rich descriptions of epistemological beliefs in the history domain, and several case studies have called the attention to the link between episte-mological beliefs and the development of historical thinking (Bain, 2000; Lee & Ashby, 2000; VanSledright, 2002b; Wineburg, 1991, 1999).

At the same time, psychological research repeatedly stressed how the beliefs people entertain about the nature, source, and the justification of knowledge influence their understanding of the information encountered in and out of school, concurring, therefore, to shape their judgment and their choices. Epistemological conceptions thus ceased to be an object of discus-sion only among philosophers and became an object of investigation also for

At the crossroads

171

educational psychologists and educational researchers (Hofer, 2002; Limon, 2002; Perry, 1970; Piaget, 1950; Pintrich, 2002).

Even if the constructs studied by these various literatures are partially overlapping, their investigations tended to follow divergent paths. The psy-chological research on epistemological beliefs has in fact been, until recently, domain-general. Moreover, when attempts were made to explore the domain-specificity of epistemological beliefs, most researchers concentrated their in-terest in the domain of science (Bell & Linn, 2002; De Corte, Eynde & Ver-schaffel, 2002; Elder, 2002). The research on historical thinking reflected the epistemological debate internal to the discipline and tended therefore to be cast in those terms. It also addressed the issues discussed among policy mak-ers and educators about the purposes of history education.

The nature of this article does not afford the opportunity for a detailed review of the literature in the areas of epistemological beliefs and historical thinking. We believe, nevertheless, that the literature in these fields can illumi-nate each other and that a fused perspective might suggest better pedagogical approaches to the teaching of history. In our view, the crossroads becomes a privileged place to grasp the trends and the pedagogical implications of these bodies of research and a fertile ground to generate new hypothesis and methods for further investigation. In particular, the review of the literature points to the emergence of a common pedagogical dilemma, which lies at the heart of epistemology. Once we leave the calm and apparently safe waters of believing that there is an unmediated and full correspondence between the external reality (the past, in the case of history) and our knowledge of it, how do we equip ourselves and our students to navigate the epistemological uncertainty that seems to ensue? How do we live up to our sowing (in Dante’s terms), avoiding at the same time the fate of Ulysses’s men whose fragile vessels were inadequate to navigate the ocean? In more modern terms, is it possible for our students to avoid intellectual drowning while traversing the thin ice of historical interpretation (VanSledright, 2002a)?

In the next sections, we gather those contributions that wrestled more closely with this dilemma. We begin by reviewing briefly the study of episte-mological beliefs in the context of cognitive research, focusing in particular on two models of epistemological development (King & Kitchener, 1994; Kuhn, 1991). Then, we look at the research on historical thinking as a par-ticularly powerful example of the influence that epistemological beliefs exert


172

on learning and teaching in a particular domain, and we compare the con-clusions of researchers working within these two traditions, exploring their intersection. Finally, we present some preliminary results of a study involving elementary teachers participating in an intensive professional development program for the teaching of American history as an example of research based on this integrated approach and an attempt at designing a measure of epistemological beliefs for the specific domain of history.

Epistemological beliefs in the context of cognitive development research

The interest of psychology in epistemological beliefs and in their de-velopment can be traced back to Piaget (1950), who referred to his inquiry into intellectual development as a study in genetic epistemology. The study of epistemology was also spurred by the exploration of Perry (1970) who sought to understand how individuals make sense of their experiences, par-ticularly with respect to formal education. Further work extended the study of epistemological beliefs to the exploration of gender differences (Baxter Magolda, 1992; Belenky, Clinchy, Goldberg, & Tarule, 1986), its relation to comprehension (Schommer, 1990), conceptual change (Dole & Sinatra, 1998; Limon, 2002), the reading of scientific texts and the subsequent development of scientific concepts (Qian & Alvermann, 1995), and the interpretation of controversial issues (Kardash & Scholes, 1996).

Some studies also suggested that culture plays a significant role in the development of epistemological beliefs (Youn, 2000) and, more generally, in the conception of knowledge and beliefs (Alexander & Dochy, 1995; Mag-gioni & Riconscente, 2003). Finally, focusing on the development of episte-mological beliefs in the academic field, researchers further investigated their domain-specificity, providing support for the hypothesis that they tend to be both domain-general and domain-specific (Buehl & Alexander, 2001; Hofer, 2000; Schommer & Walker, 1995).

Another body of research has explored the relation between teachers’ be-liefs about the epistemology of the discipline taught and beliefs about the nature of learning, teaching strategies, and students’ outcomes, further il-lustrating the importance of considering the domain-specific component of epistemological beliefs. Results highlight the close relation between teachers’

At the crossroads

173

beliefs about knowing and learning and the teaching strategies they imple-ment in the classroom (Brickhouse, 1990; Hashweh, 1996). These studies also show the effects that teachers’ beliefs about learning have on the in-clusion (or exclusion) of low-achieving students in learning activities target-ing higher-order thinking. In particular, Zohar, Degani, and Vaaknin (2001) found that teachers who shared a hierarchical view of learning tended to set different instructional goals for low and high achievers, seriously compromis-ing the learning opportunities of low achievers (Zohar & Dori, 2003; Zohar & Nemet, 2002).

For the specific purpose of this study, two approaches seem to parallel quite closely the findings of the research on historical thinking, and in par-ticular the changes often observed by the researchers. In our view, they are the Reflective Judgment (RJ) Model developed by King and Kitchener (1994), and the conceptualization of Kuhn and Weinstock (2002) of mature episte-mological understanding as “the coordination of the subjective and objective dimensions of knowing” (p.123). The opportunity to investigate epistemo-logical beliefs within a stage-like framework is one of the issues currently debated in the field. It is not clear, in fact, whether epistemological beliefs fol-low a stable, general developmental trajectory or whether they can be better conceptualized as a multifaceted, multilayered construct whose components do not necessarily develop synchronically and similarly in all the domains.

For the purpose of this study, the descriptions of the stages in King and Kitchener’s model and of the levels in Kuhn’s model have been used as char-acterizations of different epistemological stances and guided our investiga-tion of epistemological beliefs in history. The next section will therefore be dedicated to a description of the models’ features that most cover that inter-section we are interested in exploring and that have inspired the design of the measure used in the present study.

Looking for descriptions of epistemological reasoningKing and Kitchener (1994) have argued that epistemic cognition (i.e., the

process of monitoring the nature of the problem at stake, its solvability, and the evaluation of competing claims) is actually the foundation of critical thinking. That is because when individuals fail to recognize the problematic nature of the situation at stake (i.e., its uncertainty), they also tend to reduce


174

critical thinking to a set of more or less sophisticated thinking skills. The RJ Model explicitly refers to Dewey’s (1933) definition of reflective thought as “active, persistent, and careful consideration of any belief or supposed form of knowledge in the light of the grounds that support it and the fur-ther conclusion to which it tends” (p. 9). It describes, therefore, the develop-ment of reflective thinking as the changing of assumptions about the process of knowing and the justification of beliefs with individuals moving through three major periods, each one characterized by two or three stages.

For our purpose, the facets of those stages are particularly interesting, since they afford rich descriptions of how people, faced with ill-structured problems, tend to conceptualize the nature of knowledge and its justification. They offer, therefore, a useful framework to think about the development of historical thinking and provide useful hints to hypothesize how differ-ent epistemological beliefs may translate into statements about the nature of knowing, learning, and teaching history.

In the RJ Model, the first period has been named pre-reflective, to highlight that these kind of epistemic assumptions are characterized by the lack of awareness of the intrinsic uncertainty about the solution of ill-structured problems. This unawareness can manifest itself as belief that what a person observes is necessarily true and there is therefore a perfect coincidence be-tween concrete observation and knowledge. In a different way, pre-reflective thinking can give rise to the belief that it is always possible to discriminate between “right” and “wrong” knowledge. “Good” authorities know the right answers. It is also possible that pre-reflective thinking acknowledges the tem-porary uncertainty of knowledge in a few areas. In these cases, though, since knowledge is deemed impossible, beliefs are defended as a matter of personal opinion.

Quasi-reflective reasoning, the second period, is instead characterized by the realization that knowledge claims have an element of uncertainty, which can be attributed sometimes to lack of evidence or methodological problems, and sometimes to the unavoidable filtering of evidence by the person. In any case, quasi-reflective thinking is characterized by a subjective and contextual view of knowledge. Even if there is a reference to evidence, its very choice and the arguments produced are considered idiosyncratic.

Finally, reflective reasoning characterizes the third period. Reflecting think-ing is typified by the search for the most reasonable and supportable judg-

At the crossroads

175

ment, given the available evidence. Therefore, it is possible to be reasonably certain about one’s own knowledge claim, remaining at the same time open to reevaluation, if new evidence or new methods become available.

In a somewhat similar fashion, Kuhn and Weinstock (2002) identified four levels of epistemological thinking that roughly parallel the three periods iden-tified by King and Kitchener (with the first two levels considered as different possible characterization of pre-reflective thinking). While the stages of the RJ Model are mainly characterized by different justifications of knowledge, the descriptions of Kuhn and Weinstock’s levels underscore the relation be-tween knowledge and the external reality, a relation that is also crucial in the development of historical thinking. The first level, named realist, is therefore characterized by the belief that “assertions are COPIES of an external real-ity” (Kuhn & Weinstock, 2002 p. 124). In the second level, called absolutist, the possibility of false beliefs is contemplated. Assertions remain FACTS, but they can be either correct or incorrect representations of the reality. The third level, multiplist, is characterized by a vivid awareness of the role of the subject in the process of knowing and by the contemporary loss of the ob-jective dimension of knowing. Assertions become “OPINIONS freely cho-sen by and accountable only to their owners” (Kuhn & Weinstock, 2002, p. 124). Finally, at the evaluativist level, the subjective and objective dimensions of knowing are reconciled, and “assertions are JUDGMENTS that can be evaluated and compared according to criteria of argument and evidence” (Kuhn & Weinstock, 2002, p.124).

The research on historical thinking and some pedagogical dilemmas

The research in the learning and teaching of history has witnessed an im-pressive development during the past decade. In 1991, after comparing the way in which professional historians and high-school students dealt with con-trasting sources in the attempt to understand what happened at Lexington Green, Wineburg (1991) hinted at the possible connection between beliefs about historical inquiry and the strategies and heuristics typical of expertise:

For historians the picture evaluation task was an exercise in exploring the limits of historical knowledge. The end result was more a suggestion than an answer, more a forced choice from flawed alternatives than a committed decision executed with resolve. Students, on


176

the other end, responded as if the three pictures were analogous to three options on a mul-tiple-choice test. Therefore, to postulate that students believed in a single “correct answer” helps explain why they did not qualify their answers or compare the pictures with the written documents. Similarly, the use and nonuse of the sourcing heuristic could be seen as a reflec-tion of differences in each group’s conception of text, differences that addressed not just what a text says, but what a text is. (p. 83)

Even if not always expressed explicitly in the literature, examples of the connections between the beliefs about historical knowledge and disciplinary issues abound. Historical significance is a contested issue in the historical field and in curriculum, which brings to the forefront the role of the histo-rian in the making of history (Levstik, 2000). However, what happens in the classrooms is quite different. In a study with tenth-grade students, for ex-ample, Seixas (1994) found that these students seemed to move between the traditional notions conveyed by the narrative of their textbooks, and more idiosyncratic criteria that attributed historical significance only to what was significant to them as individuals. Fifth and eight graders tended to find the question about why study history at all puzzling. When pressed on the issue, they tended to espouse the criteria that we study history in order to correct our mistakes, implicitly acknowledging an understanding of history as a fixed narrative (VanSledright, 1997). The way of dealing with this issue did not change much when adults were confronted with a new understanding of what history might be. They were struck by the vividness of primary sources and complained about the dryness and impersonality of textbooks. However, once asked to write their story starting from the available evidence, adults too tended to reaffirm the objectivity of the text by seeking to be absent from their own narration (Wineburg, 1999).

Recognizing our own positionality in encountering the traces of the past is also important in order to be able to encounter new dimensions of the hu-man experience (VanSledright, 2001). This attitude, though, will be seldom cultivated if the role of the historian is not recognized (i.e., if the epistemo-logical beliefs about the nature of history do not contemplate an active role of the subject), since unchecked present standards often are used as a default to interpret the historical evidence. Wineburg (1999) reminds us that histori-cal thinking is an unnatural act; can epistemological beliefs partially account for its strangeness?

In Great Britain, the studies conducted by Peter Lee and Rosalyn Ashby

At the crossroads

177

(2000), involving students between 7 and 14 years of age, have illustrated that older students tend to view historians less as reporters and compilers, and more as active constructors of their account. Individual variability was nevertheless great, and a clean “stage-like” path was not identified. Moreover, the results of those studies also suggested that students developed at differ-ent times in different conceptual areas. A student may therefore demonstrate progression in the understanding of historical evidence while failing to show any progression in dealing with causality. Explicit academic instruction is suggested as an important factor in the development of more sophisticated views of history.

In the United States, a study conducted by Bruce VanSledright (2002a, 2002b) with fifth graders also drew attention to the role of instruction in challenging students’ epistemological stances and to the difficulties inherent in this endeavor. The method of exposing the children to the heuristics of historical investigation and the consideration of various, often conflicting sources, was found to move students from a conception of history derived from authorities or exclusively based on personal opinions to a more sophis-ticated justification of historical knowledge that takes into account the use of evidence. However, acquainting students to the idea of referring to evidence and familiarizing them with the tools of historical inquiry required a good deal of guidance and carefully scaffolded instruction. Apparently, once the absolute certainty of historical knowledge is challenged and the issue of in-terpretation openly acknowledged, it seems that students easily move from a “naïve trust to widespread suspicion” (VanSledright, 2002b, p.49). Read along the lines of epistemological development suggested by King and Kitchener, this would mean that it is relatively easy to move students within the pre-re-flective level or, in some cases, toward quasi-reflective forms of reasoning, but it is quite difficult to equip them with the tools that could favor the devel-opment of reflective thinking.

The same point is stressed by Lee and Ashby (2000), who found that students, particularly during the middle-school years, tend to believe that an opinion in history “transcends any obligation to validity or truth” (p. 222). Working with ninth graders, Robert Bain (2000) noticed that the deepening of the students’ understanding of the constructed nature of the historical narrative produced in some of them a productive skepticism. However, in others, it sustained a cynical relativism.


178

At the crossroadsThese findings highlight an important pedagogical issue. In effect, it would

be quite questionable to move students from a realist or absolutist position to a multiplist one (in Kuhn’s term) and then leave them there.

If teaching stresses distinctions between “fact” and “opinion” without making distinctions within the latter, or if it makes no attempt to differentiate provisionality from skepticism, it is likely to push adolescent views about dif-ferences in historical accounts into a potentially vicious relativism, or at least a shoulder-shrugging helplessness. Since some students have already moved beyond such positions, it would be unfortunate if teaching, in trying to of-fer helpful simplifications, appeared to justify them (Lee & Ashby, 2000, p. 222). Advocating the promotion of the intellectual tools of inquiry, analysis, and argument within the development of epistemological understanding that trains the exercise of judgment, Kuhn and Weinstock (2002) seem to echo Lee and Ashby’s (2000) statement when they observe that:

[I]t is a deceptively simple step down a slippery slope from the belief that everyone has a right to his or her opinion to the belief that all opinions are equally right. Tolerance of multiple positions, in other words, becomes confused with discriminability among them. (p. 139)

Although we believe that historical thinking involves issues which are not

exclusively epistemological in nature (Lowenthal, 2000), we also think that the available research suggests the existence of important connections be-tween historical thinking and the development of epistemological beliefs. It becomes therefore important to develop adequate measures that can enable us to further the investigation and to assess and monitor students’ beliefs change, in order to design educational intervention that can foster the acqui-sition of reflective and critical habits of mind. History education might prove to have a unique role to play in this respect.

Our aims in the present study were twofold. First, we examined a pen-and-pencil questionnaire for the evaluation of epistemological beliefs in history designed by combining the contributions of the research in epistemological beliefs and historical thinking. We also explored the compatibility of the fac-tor structure emerging from the data with the descriptions in the literature. Second, we monitored the changes in epistemological beliefs in a group of teachers involved in a professional development program for the teaching

At the crossroads

179

of American History that exposed them to the process of building historical accounts and offered pedagogical tools for implementing historical investiga-tions with fourth and fifth graders.

Some measurement issuesUp to now, epistemological beliefs have been investigated mainly through

interviews and essays. These methodologies have allowed the researchers to probe the epistemology of the participants by letting them make their own meaning. Insights into the way people reason about epistemic issues have therefore been gained. At the same time, these measures have been time-con-suming and difficult to score (Baxter Magolda, 1992; Belenky, Bond, & Wein-stock, 1997; Kitchener & King, 1981; Kuhn, 1992; Wood & Lynch, 1998).

Various alternatives that build on the Reflective Judgment Interview (RJI) have been devised. Those alternatives range from the administration of written interview questions (Wood & Lynch, 1998) to the preparation of paragraphs reflecting stage-typical responses and various paper-and-pencil questionnaires (Wood, Kitchener, & Jensen, 2002). Even if the internal con-sistency of these scales was usually in the high .70, the correlation with the data obtained through the interview (in particular the RJI) have been modest (.30−.40), suggesting that these different measures tap somewhat varied fac-ets of epistemological beliefs.

At the same time, attempts to devise paper-and-pencil measures have clar-ified some of the problems in instrument development. Written interviews, for example, have increased the chance that individuals do not engage the problem deeply or quit trying. They have also decreased the confidentiality level of the interaction between the participants and the researcher to some degree, letting issues of social desirability cloud the results. For instance, peo-ple are less likely to write that they consider their beliefs to be better than the beliefs of others, while direct interview would more easily discriminate between private beliefs and public behavior. Measures with specific reactions to a given problem have the disadvantage of forcing respondents to select from options that may not be reflective of their actual epistemological beliefs. In the case of people whose epistemological beliefs are not well articulated, this issue has been found to be particularly vexing, since specific differences among the items are not detected and a choice is made on the basis of extra-


180

neous criteria, such as the linguistic appeal of the statement.Paper-and-pencil measures targeting particular skills highlighted by the

Reflective Judgment Model have generally been more promising. Wood, Kitchener, and Jensen (2002) report the conclusions of a study exploring the differences between freshmen and seniors on a measure of correct dis-crimination between essays. Once the scores were adjusted for the effects of academic aptitude, the effect size observed with these measures was roughly the same as the one found by the RJI in comparable samples.

We believe, therefore, that, for several reasons, the study of epistemo-logical beliefs in the domain of history may profit from the development of a domain-specific, easy to administer, and objectively scorable measure. The availability of a more economical way to assess epistemic cognition may extend the exploration of the relation between epistemological beliefs and the development of historical thinking for larger samples. It can also allow the study of the relation of such beliefs with other psychological constructs, such as motivation and interest, within a particular academic domain. Finally, together with other forms of assessment, it may serve as a monitoring tool to evaluate the effects of pedagogical interventions targeting the development of historical thinking. At the same time, we are also aware that only a few aspects of epistemic cognition are tapped by this measure, which is therefore not suited to offer the rich descriptions and the subtle differentiations af-forded by qualitative research.

METHOD

ParticipantsSeventy-two fourth- and fifth-grade teachers (9 males and 63 females) par-

ticipated in a year-long professional development program. All teachers were faculty of a large school district in the mid-Atlantic region of the United States. The intervention consisted in an intensive eight-day summer course addressing issues of both content and method of teaching American history. A few after-school meetings were scheduled throughout the school year to explore further connections and applications within a revised county curricu-lum. That curriculum included several units dealing with historical investiga-

At the crossroads

181

tions. Participants were nominated by their principals. The teaching experi-ences of the participants varied greatly from 0 to 33 years (mean 10.95 years). Their previous exposure to the discipline, measured by their attendance to college level history classes, was somewhat modest (mean 3.56 classes). Five of them previously attended a class in historiography.

Epistemology measureThe Beliefs about Learning and Teaching of History questionnaire or

BLTHQ is a 21-item measure that seeks to explore domain-specific episte-mology beliefs. Respondents express their views to items on a 6-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). The complete list of the items can be found in Table 1.

Several hypotheses underlie the design of the questionnaire. First, the items attempt to mirror the stages described by King and Kitchener (2002) and are therefore presumptions of statements that would be scored as in-dicators of beliefs typical of a pre-reflective, quasi-reflective, or reflective thinking. For example, one of the characteristics of pre-reflective thinking (Stage 2) is a view of knowledge as absolutely certain and directly obtainable. In the BLTHQ, a corresponding item is: “In history there is really nothing to understand: the facts speak for themselves.” Similarly, an item constructed to mirror quasi-reflective reasoning (Stage 4) with its uncertainty and idio-syncrasy is: “Since there is no way to know what really happened in the past, students can believe whatever story they choose.” Reflective thinking (Stage 6) is tapped by items such as: “It is fundamental that students are taught to support their reasoning with evidence, and ask that history textbook do so also.”

Further, it was hypothesized that the beliefs that teachers entertain about the learning and teaching of history are an acceptable proxy of their episte-mological beliefs about history. Therefore, the statements in the BLTHQ refer to views of teaching and learning history and not directly to views of histori-cal knowledge. The reasonableness of this second hypothesis was fostered particularly by the results of the research on critical thinking and the repeat-edly emerged relation between teachers’ epistemological beliefs and teaching strategies (Brickhouse, 1990; Hashweh, 1996; Lipson, Mosenthal, Daniels, & Woodside-Jiron, 2000; Zohar, Degani, & Vaaknin, 2001). Given the nature of


182

our study, we also thought that this formulation would prove more familiar to elementary teachers, who are typically not disciplinary specialists.

Item Factor

1 2

Students who are good at memorization learn history quickly. .247 .0014

Corroborating evidence and identifying sources are important learning strategies in history, but only after mastering the basic facts.

.472 .020

In history there is really nothing to understand; the facts speak for themselves. .633 .223

Knowledge of the historical method is fundamental for historians and students alike. -.291 -.259

Students who know their textbook well will be good at history. -.013 .080

To learn history means mainly to study many facts about the past and commit them to memory.

.576 .030

Teachers need to avoid giving students conflicting sources, since it makes historical inves-tigation impossible.

.718 -.043

In learning history, summarizing is more important than comparing. .543 -.043

Teachers should not question students’ historical opinions, only check that they know the facts.

.518 -.158

Good general reading and comprehension skills are enough to learn history well. .360 -.044

Comparing sources and looking for author subtext are essential components of the process of learning history.

-.712 .060

Students need to be taught to deal with conflicting evidence. -.450 -.054

It is fundamental that students are taught to support their reasoning with evidence and ask that history textbook authors do so also.

-.553 -.099

Students who read many history books learn that the past is what the historian makes it to be.

-.347 .545

Good students know that history is basically a matter of opinion. .164 .657

Students need to be aware that history is essentially a matter of interpretation. -.219 .576

History should be taught like a story: Some things are true, but some others are just a matter of personal opinion.

-.054 .571

In reading a history book, it is more important to pay attention to the perspective of the historian than to his/her reasoning on the evidence discussed.

-.180 .587

Since there is no way to know what really happened in the past, students can believe what-ever story they choose.

.276 .595

Teaching that one historical interpretation is better than another is usually inappropriate. .175 .305

Teachers need to make all historical interpretations available and let the students construct their own understanding of them.

.080 .230

TABLE 1. Two-Factor Solution for the BLTHQ at Pre-Intervention

At the crossroads

183

ProcedureThe BLTHQ was administered as part of a battery of measures. The first

administration occurred during a preliminary meeting, two months prior the summer course. It was administered a second time 4 months later, which was 2 months after the conclusion of the summer intervention.

RESULTS AND CONCLUSIONS

Exploratory Factor AnalysisPrincipal component analysis with Varimax rotation was performed sepa-

rately on the pre- and post-intervention questionnaires. In both cases, analy-sis of the scree plot suggested the extraction of two principal components. The percentage of the total variance explained was 28.9 for the first set of questionnaires and 32.71 for the second set. The loadings in the Rotated Component Matrix are reported in Table 1 for the pre-intervention and Table 2 for the post-intervention. Using a loading of .30 as a decision point, we interpreted the first factor as indicative of an overall pre-reflective stance. This factor was perceived as underlying beliefs in the fixed and exclusively objective nature of history, a view that tends to forget the significance of the historian and of the historical method in the development of historical knowledge. It included items such as: “To learn history means mainly to study many facts about the past and commit them to memory;” and, “In learning history, summarizing is more important than comparing.” It also included, but with negative loadings, items created to mirror reflective thinking such as: “Comparing sources and looking for author subtext are essential compo-nents of the process of learning history;” and, “Knowledge of the historical method is fundamental for historians and students alike.”

Based on the loadings, we interpreted the second factor to represent qua-si-reflective reasoning. In effect, this factor was seen as indicative of beliefs in the fundamentally subjective nature of history, a view that tends to equate interpretation with opinion. Items that loaded on this factor included: “Stu-dents who read many history books learn that the past is what the historian makes it to be;” and, “History should be taught like a story: Some things are true, but some others are just a matter of personal opinion.”


184

Item Factor

1 2

Students who are good at memorization learn history quickly. .460 .098

Corroborating evidence and identifying sources are important learning strategies in his-tory, but only after mastering the basic facts.

.371 .308

In history there is really nothing to understand; the facts speak for themselves. .329 .210

Knowledge of the historical method is fundamental for historians and students alike. -.692 .183

Students who know their textbook well will be good at history. .359 .193

To learn history means mainly to study many facts about the past and commit them to memory.

.465 .230

Teachers need to avoid giving students conflicting sources, since it makes historical in-vestigation impossible.

.324 .247

In learning history, summarizing is more important than comparing. .529 -.081

Teachers should not question students’ historical opinions, only check that they know the facts.

.776 -.166

Good general reading and comprehension skills are enough to learn history well. .387 -.070

Comparing sources and looking for author subtext are essential components of the pro-cess of learning history.

-.704 .147

Students need to be taught to deal with conflicting evidence. -.550 .036

It is fundamental that students are taught to support their reasoning with evidence and ask that history textbook authors do so also.

-.683 .067

Students who read many history books learn that the past is what the historian makes it to be.

-.068 .630

Good students know that history is basically a matter of opinion. .074 .638

Students need to be aware that history is essentially a matter of interpretation. .039 .689

History should be taught like a story: Some things are true, but some others are just a matter of personal opinion.

-.34 .692

In reading a history book, it is more important to pay attention to the perspective of the historian than to his/her reasoning on the evidence discussed.

-.174 .575

Since there is no way to know what really happened in the past, students can believe whatever story they choose.

.058 .616

Teaching that one historical interpretation is better than another is usually inappropriate. .043 .346

Teachers need to make all historical interpretations available and let the students construct their own understanding of them.

-.286 .262

TABLE 2. Two-Factor Solution for the BLTHQ at Post-Intervention

Even if the value of specific item loadings changed from pre- to post-in-tervention, their sign and overall structure did not. After reverse coding items with negative loadings and dropping one item from both sets whose loading

At the crossroads

185

was smaller than .30, a scale was built by adding the row scores of each par-ticipant on the individual items. A high score on Factor 1

(M = 28.03, SD = 5.33 at pre-treatment and M = 24.60, SD = 5.81 at post-treatment) was interpreted as an overall agreement with beliefs typical of pre-reflective thinking and a high score on Factor 2 (M = 24.23, SD = 4.47 at pre-treatment and M = 25.42, SD = 5.59 at post-treatment) was interpreted as an overall agreement with beliefs typical of quasi-reflective thinking. Alter-natively, a low score on Factor 1 could be interpreted as an overall agreement with beliefs indicative of a reflective reasoning. It was actually this second interpretation that better fit the scores of the participants in our study. The reliability of the scale for Factor 1, measured by Cronbach alpha, was .72 for both sets of data. The reliability of the scale for Factor 2 was .63 for the first set and .70 for the second set.

Two further analyses were performed on the data. A two-factor principal component analysis with Oblimin rotation yielded the same two-factor struc-ture and a correlation of .04 among the factors, strongly suggesting their or-thogonality. A three-factor solution was also explored. However, the overall theoretical consistency of the three-factor solution was weaker than the one afforded by a two-factor solution. The reliability of the scale for the third fac-tor was also low (α = .27 for the first set of data and α =.57 for the second).

The factor structure that emerged from the analysis was in some measure surprising, if considered in light of the literature on epistemological beliefs. Specifically, both the items mirroring pre-reflective and reflective thinking loaded together on Factor 1, albeit with opposite sign. By comparison, items indicative of quasi-reflective thinking plainly loaded on Factor 2. This occur-rence may be considered puzzling, particularly if pre-reflective and reflective thinking are conceptualized as opposite extremes along a developmental tra-jectory.

Yet, if the factor structure is considered in relation to the domain of histo-ry, it seems more comprehensible. That is, this structure, in our view, reflects the very tension characterizing the generation of historical knowledge. On one hand, historians look at the past through the lenses of their own ques-tions and positionalities. Therefore, they cannot avoid recognizing, in some measure, the subjective aspect of the knowledge so generated. On the other hand, the available evidence sets the boundaries of their search, suggesting new avenues of investigation and new hypotheses of explanations, and thus


186

grounding their work in an external reality. Thus, one way to represent the various epistemological stances is to con-

ceptualize them along the two orthogonal dimensions emerged from the ex-ploratory factor analysis. Compared to a linear representation of epistemo-logical development, we believe that this bi-dimensional conceptualization can better capture epistemological beliefs in the domain of history, offer-ing a better accommodation to the broad range of epistemological beliefs that also characterize the stances of professional historians. It would also be consistent with a phenomenon often observed by researchers in historical thinking (Bain, 2000; Lee & Ashby, 2000; VanSledright, 2002a, 2002b), i.e., the facility with which novice students, once exposed to conflicting historical sources and alternative interpretations, make the epistemological jump into a resigned naïve relativism. In other words, it is not uncommon that individuals who are used to think of history as coinciding with the past, once exposed to conflicting or lacking evidence tend to conclude that, whenever absolute objectivity is impossible, history becomes, for the most part, just a matter or personal opinion. Thus, they tend to manifest some of the traits typical of pre-reflective thinkers coupled with some of the characteristics of quasi-re-flective thinkers.

Mapping epistemological shiftsDifferences between the means at pre- and post-treatment were signifi-

cantly different on both factors (t = 5.171 for Factor 1 and t = -2.255 for Factor 2). At post-treatment, teachers scored, on average, lower on Factor 1 and higher on Factor 2. On average, after the intervention, the degree of agreement with items portraying reflective thinking increased. At the same time, teachers also tended to agree more (or disagree less) with items indica-tive of quasi-reflective thinking.

In order to understand the epistemological shift better, we decided to fo-cus the analysis on individual teachers by building epistemological profiles. Mid-points of the scales were calculated to separate areas of average agree-ments with the statements reflecting the factors from areas of average dis-agreement. Since 13 items formed the scale for Factor 1, the mid-point for Factor 1 was 45.5. The scale for Factor 2 comprised 7 items and its mid-point was therefore 24.5.

At the crossroads

187

According to the score obtained on the two factors, four different combi-nations could have theoretically occurred (i.e., below the mid-point on both factors; above the mid-point on both factors; above the mid-point on one factor but below on the other). These four potential profiles are represented in Table 3.

Criterialists Naïve Realists

Relativists Dichotomous Thinkers

F A C T O R 1

TABLE 3. Mapping Epistemological Profiles

Factor 1 serves as a proxy for beliefs in the objective aspect of historical knowledge, while Factor 2 is a proxy for beliefs in its subjective aspect. The profiles so generated are characterized by different ways of reconciling the subjective and objective dimensions of knowing. In terms of the BLTHQ, reflective thinking is best represented by relatively low scores on both the pre-reflective and quasi-reflective factors. We have termed this position criterialist, to underline the search for the best explanation through the patient weav-ing together of the best evidence and the best argument available. Relatively high scores on Factor 2 and low scores on Factor 1, indicate a relativist posi-

FACTOR

2

Low

High

Low High


188

tion, epitomized by the emphasis placed on the subjective nature of historical knowledge. In this case, the constructed nature of history is acknowledged, but the historical method is not deemed an effective tool to deal with prob-lems of conflicting or missing evidence. The upper right quadrant, where scores are high on Factor 1 and low on Factor 2, signals what we would term a naïve realist position, characterized by the belief in a perfect correspondence between the past and history. In this case, focus on the evidence fogs the awareness of the interpretive process at work. We would speculate that these beliefs tend to be typical of novices, who still do not possess a well developed knowledge of the principles and methods that characterize a specific domain (Alexander, 2003). Finally, the bottom right quadrant, with high scores on both factors, signals in our view a form of dichotomous thinking, where beliefs in the unmediated nature of historical knowledge accompany a view of his-tory as prevalently subjective.

Since none of our participants had a score on Factor 1 higher than 45.5, only two profiles actually emerged from our data. Our sample was composed of teachers, and the BLTHQ items were framed in terms of beliefs about the teaching of history. It is possible, therefore, that preferences for a constructiv-ist teaching style, usually considered socially desirable, concur to explain this occurrence. Considering that the BLTHQ has not been validated so far in any other sample, an alternative explanation would be that this measure is not ad-equate to tap pre-reflective beliefs. Further research, with a more diversified sample, is necessary to explore these possibilities. Following the theoretical profiles, we assigned teachers to one of the following two groups: criterialist or relativist. A summary of teachers’ shifts from pre- to post-treatment is sum-marized in Figure 1.

In particular, on the pre-treatment, 37 teachers were classified as criterialists. On the post-treatment, 24 of them were still in this group, while13 moved to the relativists group. The relativist group on the pre-treatment counted 35 teachers. On the post-treatment, 5 of these moved to the criterialist group, while 30 remained in the relativist group.

Trying to better understand these movements, we focused our analysis on the teachers that moved across groups after the treatment and investigated whether the degree of their agreement with statements indicative of reflec-tive thinking was relatively strong or weak. The five teachers that moved from the relativist to the criterialist group all scored on the pre-treatment relatively

At the crossroads

189

high on Factor 1 (scores higher than 22.75, first quartile of the scale for Fac-tor 1).

FIGURE 1. Changes in Teachers’ Epistemological Beliefs

Bordering on almost dichotomous thinking, these teachers were more in-clined to believe in the unmediated nature of historical knowledge, although they tended to equate historical interpretation with opinion, at the same time. Eleven of the 13 teachers that moved from the criterialist to the relativist group also tended to score high on Factor 1 at pre-treatment. Nine of these teach-ers still scored relatively high on Factor 1 at post-treatment, simultaneously expressing some beliefs in the unmediated nature of history and in its sub-stantial subjectivity.

DISCUSSION

A few preliminary considerations on the limitations of the study are nec-essary to qualify the reflections on this initial exploration of the crossroads between the literatures in epistemological beliefs and historical thinking. In particular, two features of the study need to be addressed. First, the charac-teristics of the sample, composed by fourth and fifth grade teachers, limit the generalizability of the results. However, this feature makes these outcomes

Criterialist37

Criterialist29

Relativist35

Relativist43

24

13

30

5

Pre-treatment Post-treatment


190

particularly interesting, since very few studies, in our knowledge, have ad-dressed elementary school teachers’ epistemologies in the domain of history. Second, although the BLTHQ is a preliminary attempt to design a measure for studying epistemological beliefs in the domain of history, more research is needed to refine it and adapt to the investigation of different developmental ages. Further, only a relatively small part of the total variance was explained by the two factors underlying the scales. Although this is a common occur-rence in the literature and the reliability of the scales is acceptable, it suggests that much of the epistemological beliefs construct still eludes measurement.

Nevertheless, we believe that the exploration of the crossroads remains fruitful, offering a more articulate framework for research design and inter-pretation. In particular, from a methodological standpoint, it fostered the design of a domain-specific measure of epistemological beliefs, a step neces-sary to further our understanding of the development of historical thinking. From a broader theoretical point of view, it suggested conceptualizing the epistemological development in the domain of history along the two dimen-sions of the objective and subjective aspects of historical knowledge, a view that guided the construction of epistemological profiles and afforded an ini-tial way to map epistemological change.

The epistemological shifts we observed after the participation of the teachers in the summer program seem to suggest a relative stability of epis-temological beliefs in time. This occurrence appears to confirm that episte-mological beliefs need to be explicitly targeted and that explicit instruction in and repeated exposure to the heuristics typically employed by historians is necessary to challenge epistemological views in history. It is quite plau-sible that the summer program offered to these teachers was not enough to promote a revision of their previous beliefs and that more time is required to foster change. This article does not afford an adequate examination of the professional development program that took place between the two ad-ministrations of the BLTHQ. Details about the program, its results, and its implications are provided in VanSledright, Alexander, Maggioni, Kelly and Meuwissen (2004).

However, even if moderate, the shifts we observed indicate that individual movements across different epistemological positions are not unidirectional, supporting the hypothesis that progression in historical thinking does not follow a rigid, stage-like path. At the same time, it seems that shifts toward

At the crossroads

191

beliefs highlighting the subjective nature of knowledge were more common than shifts signaling a reconciliation of the objective and subjective aspects of it. The fact that these shifts generally involved participants who tended to border some form of dichotomous thinking may also suggest that individu-als tend to revert to this kind of reasoning when their epistemological beliefs undergo change. Finally, the hypothesis of the multidimensional nature of epistemological beliefs is compatible with these results. Of course, such hy-potheses require more direct empirical testing.

The mapping of the epistemological shifts indicated that roughly half of the teachers in our sample fell in the relativist quadrant and, after the summer program, the number of teachers assigned to this profile increased. This re-sult recalls the pedagogical dilemma described in the literature and it appears particularly worrisome since this kind of beliefs is usually at odds with the development of historical thinking. How will teachers be able to coach their students in this kind of thinking, if they had no opportunity to familiarize with its tenets? Given the centrality of their role in the education process, we find this issue even more pressing, since no curricula can substitute the daily, living relationships between students and their teacher, especially when the target is some form of higher level thinking.

We believe that this problem may be addressed at least at two levels. The first regards the education and training of pre-service teachers, who need to be exposed and equipped to traverse the thin ice of historical interpreta-tion. A reflection on the epistemological foundations of the various subject matters becomes therefore fundamental and implies an approach to the dis-ciplines not limited to the acquisition of content knowledge but extended to the awareness of how knowledge in a particular field is generated in the first place. It is difficult to understand why this opportunity is reserved, at best, to high-school teachers, since there is no developmental evidence that children are unable to handle sophisticated epistemological thinking, at least in the domain of history. However, once ingrained, epistemological beliefs are usu-ally hard to change. For these reasons, setting the goal of historical thinking only for older or high-achieving students becomes, in our view, even more questionable.

The second level involves the education of in-service teachers. In the spe-cific case of history, research would predict that exposing students to the conflictual nature of evidence without at the same time providing them with


192

the tools to deal with this tension is an almost sure recipe for fostering naïve relativism and cynical skepticism. If the goal is teaching students to think critically, then we believe that an investment in professional development be-comes necessary. Thinking historically and implementing strategies that foster historical thinking among the students is a difficult task, a task that requires the development of several skills bordering domain-expertise. Teachers will need education, tools, and some form of continuous support to pursue these educational goals. Case studies also suggest that an explicit discussion of the epistemological foundation of the discipline, together with opportunities to familiarize teachers with the heuristics and way of thinking of professional historians might foster the development of teaching strategies that favor historical thinking (Kelly, Meuwissen, Maggioni, VanSledright & Alexander, 2004.) Further research devoted to better understand how shifts in epistemo-logical thinking in specific domain occur and to investigate whether patterns exist is needed.

Historical thinking can foster inquiry, investigation, and debate, provid-ing students with standards and tools to participate more fully in the con-temporary culture, as members of a liberal, open society. More importantly, perhaps, it can promote the development of a culture in which the awareness of the limits of our knowledge, far from dooming our reason, sustains our curiosity and our quest.

REFERENCES

Alexander, P. A. (2003). The development of expertise: The journey from acclimation to proficiency. Educational Researcher, 32(8), 10-14.

Alexander, P. A., & Dochy, F. J. R. C. (1995). Conceptions of knowledge and beliefs. A comparison across varying cultural and educational communi-ties. American Educational Research Journal, 32, 413-442.

Alighieri, Dante (1996). The Divine Comedy of Dante Alighieri. Edited and trans-lated by Robert M. Durling. Volume 1, Inferno. New York: Oxford Univer-sity Press.

Bain, R. B. (2000). Into the breach: Using research and theory to shape his-tory instruction. In Stearns, P. N., Seixas, P., & Wineburg, S. (Ed.), Knowing, teaching and learning history: National and international perspectives (pp. 331-352).

At the crossroads

193

New York University Press.Baxter Magolda, M. (1992). Knowing and reasoning in college: Gender-related patterns

in students’ intellectual development. San Francisco: Jossey-Bass.Belenky, M. F., Bond, L. A., & Weinstock, J. S. (1997). A tradition that has no

name: Nurturing the development of people, families, and communities. New York: Basic Books.

Belenky, M. F., Clinchy, B. M., Goldberg, N. R., & Tarule, J. M. (1986). Wom-en’s ways of knowing: The development of self, voice, and mind. New York: Basic Books.

Bell, P., & Linn, M. C. (2002). Beliefs about science: How does science in-struction contribute? In Hofer, B. K. & Pintrich, P. R. (Eds.), Personal epis-temology: The psychology of beliefs about knowledge and knowing (pp. 297-320). Mahwah, NJ: Lawrence Erlbaum Associates.

Brickhouse, N. W. (1990). Teachers’ beliefs about the nature of science and their relationship to classroom practice. Journal of Teacher Education, 41(3), 53-62.

Buehl, M. M., & Alexander, P. A. (2001). Beliefs about academic knowledge. Educational Psychological Review, 13, 385-418.

De Corte, E., Eynde, P. O., & Verscshaffel, L. (2002). “Knowing what to believe”: The relevance of students’ mathematical beliefs for mathematics education. In Hofer, B. K. & Pintrich, P. R. (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 297-320). Mahwah, NJ: Lawrence Erlbaum Associates.

Dewey, J. (1933). How we think: A restatement of the relation of reflective thinking to the

educative process. Boston: D.C. Heath and Company.Dole, J. A., & Sinatra, G. M. (1998). Reconceptualizing change in the cogni-

tive construction of knowledge. Educational Psychologist, 33, 109-128.Elder, A. D. (2002). Characterizing fifth-grade students’ epistemological be-

liefs in science. In Hofer, B. K. & Pintrich, P. R. (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 347-363). Mahwah, NJ: Lawrence Erlbaum Associates.

Hashweh, M. (1996). Effects of science teachers’ epistemological beliefs in teaching. Journal of Research in Science Teaching, 33(1), 47-63.

Hofer, B. K. (2000). Dimensionality and disciplinary differences in personal epistemology. Contemporary Educational Psychology, 25, 378-405.


194

Hofer, B. K. (2002). Personal epistemology as a psychological and educa-tional construct: An introduction. In Hofer, B. K. & Pintrich, P. R. (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 3-14). Mahwah, NJ: Lawrence Erlbaum Associates.

Kardash, C. M., & Scholes, R. J. (1995). Effects of preexisting beliefs and repeated readings on beliefs change, comprehension, and recall of persua-sive text. Contemporary Educational Psychology, 20, 201-221.

Kelly, T., Meuwissen, K, Maggioni, L., VanSledright, B., & Alexander, P. A. (2004, April). Exploring changes in historical thinking and pedagogy: Comparative case studies of three elementary teachers. Poster to be presented at the Annual Meeting of the American Educational Research Association. San Diego, CA.

King, P. M. & Kitchener, K. S. (1994). Developing reflective judgment: Understand-ing and promoting intellectual growth and critical thinking in adolescents and adults. San Francisco: Jossey-Bass Publishers.

King, P. M., & Kitchener, K. S. (2002). The Reflective Judgment Model: Twenty years of research on epistemic cognition. In Hofer, B. K. & Pin-trich, P. R. (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 37-61). Mahwah, NJ: Lawrence Erlbaum Associates.

Kitchener, K. S., & King, P. M. (1981). Reflective judgment: Concepts of justification and their relationship to age and education. Journal of Applied Developmental Psychology, 2, 89-116.

Kuhn, D. (1991). The skills of argument. Cambridge University Press.Kuhn, D. (1992). Thinking as argument. Harvard Educational Review, 62, 155-

178.Kuhn, D. (1999). A developmental model of critical thinking. Educational Re-

searcher, 28, 16-25, 46.Kuhn, D., & Weinstock, M. (2002). What is epistemological thinking and why

does it matter? In Hofer, B. K. & Pintrich, P. R. (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 121-144). Mahwah, NJ: Lawrence Erlbaum Associates.

Lee, P. & Ashby, R. (2000). Progression in historical understanding among students ages 7-14. In Stearns, P. N., Seixas, P., & Wineburg, S. (Eds.), Knowing, teaching and learning history: National and international perspectives (pp. 199-222). New York University Press.

Levstik, L. S. (2000). Articulating the silences: Teachers’ and adolescents’

At the crossroads

195

conceptions of historical significance. In Stearns, P. N., Seixas, P., & Wine-burg, S. (Eds.), Knowing, teaching and learning history: National and international perspectives (pp. 284-305). New York University Press.

Limon, M. (2002). Conceptual change in history. In Limon, M. & Mason, L. (Eds.), Reconsidering conceptual change: Issues in theory and practice (pp. 259-289). Dordretch, The Netherlands: Kluwer Academic Press.

Lipson, M., Mosenthal, J., Daniels, P., & Woodside-Jiron, H. (2000). Process writing in the classroom of eleven fifth-grade teachers with different ori-entations to teaching and learning. The Elementary School Journal, 101(2), 209-231.

Lowenthal, D. (2000). Dilemmas and delights of learning history. In Stearns, P. N., Seixas, P., & Wineburg, S. (Eds.), Knowing, teaching and learning his-tory: National and international perspectives (pp. 63-82). New York University Press.

Maggioni, L, & Riconscente, M. (2003, August). Perceptions of knowledge and beliefs among undergraduate students in Italy and in the United States. Paper presented at the 10th Biennial Conference of the European Associa-tion for Research on Learning and Instruction, Padova, Italy.

Novick, P. (1988). That noble dream: The “objectivity question” and the American historical profession. Cambridge University Press.

Perry, W. G. (1970). Forms of intellectual and ethical development in the college years: A scheme. New York: Holt, Rinehart, and Winston.

Phillips, D. C. (1995). The good, the bad, and the ugly: The many faces of constructivism.

Educational Researcher, 24(7), 5-12.Piaget, J. (1950). Introduction a l’epistemologie genetique. Paris: Presses Univ. de

France.Pintrich, P. R. (2002). Future challenges and directions for theory and re-

search on personal epistemology. In Hofer, B. K. & Pintrich, P. R. (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 389-414). Mahwah, NJ: Lawrence Erlbaum Associates.

Qian, G., & Alvermann, D. (1995). Role of epistemological beliefs and learned helplessness in secondary school students’ learning science concepts from text. Journal of Educational Psychology, 87(2), 282-292.

Schommer, M. (1990). Effects of beliefs about the nature of knowledge on comprehension. Journal of Educational Psychology, 82, 498-504.


196

Schommer, M., & Walker, K. (1995). Are epistemological beliefs similar across domains? Journal of Educational Psychology, 87(3), 424-432.

Seixas, P. (1994). Students’ understanding of historical significance. Theory and Research in Social Education, 23(3), 281-304.

Seixas, P. (2000). Schweigen! Die Kinder! Or does postmodern history have a place in the schools? In Stearns, P. N., Seixas, P., & Wineburg, S. (Ed.), Knowing, teaching and learning history: National and international perspectives (pp. 19-37). New York University Press.

Stearns, P. N., Seixas, P., & Wineburg, S. (2000). Knowing, teaching and learning history: National and international perspectives. New York University Press.

VanSledright, B. (1997). And Santayana lives on: Students’ views on the pur-poses for studying American history. Curriculum Studies, 29(5), 529-557.

VanSledright, B. (2001). From empathic regards to self-understanding: Im/positionality, empathy, and historical contextualization. In Davis Jr., O. L., Yeager, E. A., & Foster, S. J. (Eds), Historical empathy and perspective taking in the social studies (pp. 51-68). Lanham, MD: Rowan Littlefield.

VanSledright, B. (2002a). Confronting history’s interpretive paradox while teaching fifth graders to investigate the past. American Educational Research Journal, 39, 1089-1115.

VanSledright, B. (2002b). In search of America’s past: Learning to read history in elementary school. New York: Teachers College Press.

VanSledright, B., Alexander, P. A., Maggioni, L., Kelly, T., & Meuwissen, K. (2004, April). Examining shifts in teachers’ epistemologies in the domain of history. Paper to be presented at the Annual Meeting of the American Educational Research Association. San Diego, CA.

von Glasersfeld, E. (1991). Radical constructivism in mathematics education. Dor-drecht, The Netherlands: Kluwer Academic Publishers.

Wineburg, S. (1991). Historical problem solving: A study of the cognitive processes used in the evaluation of documentary and pictorial evidence. Journal of Educational Psychology, 83(1), 73-87.

Wineburg, S. (1999, March). Historical thinking and other unnatural acts. Phi Delta Kappan, 489-499.

Wood, P., Kitchener, K., & Jensen, L. (2002). Considerations in the design and evaluation of a paper-and-pencil measure of epistemic cognition. In Hofer, B.K. & Pintrich, P.R. (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 277-294). Mahwah, NJ: Lawrence

At the crossroads

197

Erlbaum Associates.Wood, P.K., & Lynch, C. (1998). Using guided essays to assess and encourage

reflective thinking. Assessment Update, 10, 14-15.Youn, I. (2000). The cultural specificity of epistemological beliefs about

learning. Asian Journal of Social Psychology, 3, 87-105.Zohar, A, Degani, A., & Vaaknin, E. (2001). Teachers’ beliefs about low-

achieving students and higher order thinking. Teaching and Teacher Education, 17, 469-485.

Zohar, A., & Dori, Y. J. (2003). Higher order thinking skills and low-achieving students: Are they mutually exclusive? The Journal of the Learning Sciences, 12(2), 145-181.

Zohar, A., & Nemet, F. (2002). Fostering students’ knowledge and argumen-tation skills through dilemmas in human genetics. Journal of Research in Science Teaching, 39(1), pp. 35-62.



European Journalof

School Psychology

Articles

European Journal of School Psychology - Vol. 2 , No. 1-2, 201-218 (2004)

201

CREATING NEW CLASSROOM CULTURE VIA A VIRTUAL WORLD:

A CASE STUDY

Henny van der Meijden*1, Maria Beatrice Ligorio**

THE CASE OF IVO

In the present study, the effects of the construction of a three-dimensional virtual world using CSCL software on the learning and development of a pupil diagnosed with a form of autism are considered. The pupil, Ivo, was in ninth grade when a project devoted to the creation of a virtual world was initiated at his Montessori school in The Hague, The Netherlands, in cooperation with other Dutch and Italian schools. One of Ivo’s teachers noticed considerable computer skill and interest on the part of Ivo and therefore encouraged him to participate in the project. Ways to include Ivo as a productive and useful member of the project were sought, and it was expected at the same time that Ivo’s participation in the project would help him overcome both his learning and personal problems. Ivo was found to participate in the project in a pro-gressively independent and fruitful manner and to even serve as an on-line tutor for other pupils with little computer experience in the end.

TECHNOLOGY AND LEARNING PROBLEMS

For pupils with learning problems, Information Computer Technology (ICT) can play an important role in their education. That is, ICT is increas-

*University of Nijmegen (NL)**University of Bari (IT) 1Corrispondance concerning this article should be addressed to Henny van der Meijden, Department of Educa-tional Sciences, University of Nijmegen – PO Box 9104 -6500 HE Nijmegen. The Netherlands; Tel. Int. +31 24 3612585. Email: [email protected]

Henny van der Meijden, Maria Beatrice Ligorio

202

ingly being used in many countries and in different contexts to enhance not only the learning of normally progressing pupils but also pupils experiencing difficulties.

In fact, ICT appears to have an enormous potential to reduce the learning difficulties and compensate for many of the disabilities experienced by pupils with learning problems.

Support and help can be provided in many different ways as the range of available ICT resources is quite wide and such diversity can obviously help meet the special educational needs of pupils in a variety of manners. Listed below are only a few of the educational advantages which most ICT resources can provide.

a) Reporting and writing. Word processors are commonly used to store and process written text. Some pupils may find it is easier to write using a computer than by hand and therefore benefit from the use of a stand-ard word processor. Such word processors often provide facilities for the checking of spelling and syntax in general and particularly when the visual perception of the user is impaired. Furthermore, the correction of computer-written texts is much easier and less time-consuming than the correction of hand-written texts. ICT thus supports the production of high quality documents.

b) Motivation. A neatly printed copy of something tends to be much more satisfying and motivating than a hand-drafted copy. Mistakes are quick-ly corrected on a computer, and users can quickly proceed to more complex or challenging tasks. It is also being increasingly recognized that the use of computers in education and particularly to interact with peers in remote locations is very, very motivating. Learning problems can obviously affect pupil motivation and involvement in a task. And motivation has been found to constitute a particularly acute problem for pupils with special needs as they have generally not had the experi-ence of academic investment improving their lives (Goldenberg, Rus-sell, Carter, 1984).

c) Personal development. With the use of ICT, pupils can devote as much time as they want or need to the accomplishment of a task. Privacy is preserved because a pupil only makes his or her work available to others when he or she wants to. That is, ICT provides a ‘safe’ learning

Creating new classroom culture via a virtual world: a case study

203

environment where new skills can be acquired and developed without others seeing failures which may occur along the way. There is time for reflection. Pupils can thus concentrate on their thoughts and ideas and only enter them into the computer when they feel confident; think about an answer before responding to a question; read and re-read texts; and focus on the sequencing of thoughts and ideas — which fosters the development of metacognitive skills.

d) Communication. Computers can provide support for children who have communication difficulties. For particularly children with poor ver-bal capacities and children who are too shy to speak, ICT provides a number of alternative communication formats. For instance, multime-dia formats often decrease the quantity of text to be read and provide more space for symbols, pictures and sounds — both real and syn-thetic. All of the foregoing are easily combined, and pupils with com-munication and/or reading difficulties can thus find a format which clearly fits their needs.

e) Interaction and collaboration within learning networks. Computer-based ac-tivities and games can be used by two or more children to encourage interaction and collaboration at both the keyboard and a distance. Vari-ous social, affective and cognitive skills are greatly stimulated in such a manner (Caravita & Ligorio, 2003). Learning tasks are made fun, and children are thereby encouraged to concentrate for long periods of time. Collaboration with others via computers (Computer Supported Collaborative Learning, CSCL) has proved most motivating and in-structive (Talamo, Nimivierta, 2000).

In fact, one of the main goals of CSCL is to improve interpersonal communication processes via computers based on the assumption that the efficacy of a group largely depends on just how the group mem-bers communicate with each other. CSCL when uses distance com-munication provides numerous opportunities for those pupils who are normally inhibited during face-to-face interaction. That is, successful participation in non-mediated collaborative groups is often competi-tive and requires speed, alertness, self-confidence and a variety of con-versational skills. In CSCL groups, in contrast, interruption and atten-tion-getting skills play little or no role (Van der Meijden, Veenman,


204

in press). Status differences typically disappear during CSCL and par-ticipation is often more equal than in face-to-face interaction (Bordia, 1997). CSCL thus gives pupils the opportunity to demonstrate skills which they might otherwise never demonstrate, and CSCL can thus be used to integrate pupils into the school and a wider community.

Consideration of the use of ICT resources for educational purposes shows them to be very promising and effective when adequate pedagogical models are available to guide their application (diSessa, 2000).

Studies of CSCL have shown computers to promote the use of collabora-tive strategies in many ways, which is a very important finding in light of the fact that collaborative learning is currently considered critical for not only pu-pil achievement but also their social development (Dillenbourg, 1999; John-son, Johnson, 1986; Slavin, 1990).

AUTISM AS A PARTICULAR LEARNING PROBLEM

Autism is a serious developmental disorder (American Psychiatric Asso-ciation, 1987).

Sensory input to the brain is incorrectly perceived or interpreted, which can result in a confused understanding of life and the environment, strange behaviour, withdrawal, perceptive and cognitive disorders and retardation (Baron-Cohen, Tager-Flusberg & Cohen, 2000).

All of the symptoms are not necessarily present in every autistic child, and the severity of the symptoms can vary greatly. People with autism typically display the so-called Triad of Impairments or problems in three areas:

• social interaction;• communication;• behaviour (i.e., restricted interests and repetitive movements). Communication and language problems are a major feature of autism

with more than 50% of autistic people never developing the capacity for meaningful communication and most people with autism experiencing com-munication problems.

Autistic individuals are easily distracted and have poor organizational skills. They may also find it difficult to deal with abstractions and become


205

obsessed with details. And it is obvious that such communication problems can further hamper the learning and social development of autistic individu-als (Wetherby, Prizant, 2000).

Many people with an Autism Spectrum Disorder also experience sensory sensitivities. They may be over or under sensitive within a number of do-mains, including those of sight, touch, taste, smell, sound, temperature and pain. As a result of these impairments, people with an autistic disorder may also experience overwhelming anxiety, frustration and confusion when faced with the demands of everyday life.

The majority of people with an autistic disorder also have an intellectual disability although some of them may have average or above-average intel-ligence and relatively good communication skills with specific learning dis-abilities.

Although Autism Spectrum Disorders are almost always present at birth, they often do not become evident until the age of 2 or 3 years. Studies sug-gest that Autism Spectrum Disorders are about four times more prevalent in boys than in girls and that 1 out of 100 people may have some form of the disorder. Autism is often associated with other neurological disorders and therefore difficult to diagnose.

Close observation of a child’s behaviour can lead to a diagnosis, but au-tism remains a lifelong disability and therefore affects the manner in which the individual relates to other people and the world around her/ him. Con-crete behavioural indicators of autism are: impaired communication (e.g., dif-ficulties with the interpretation and use of gestures); flat facial expression and vocal tone; repetition of heard phrases and words; severely impaired social skills (e.g., difficulties with the maintenance of eye contact, difficulties initiat-ing and sustaining a conversation); very specific and unusual preoccupations; and a preference for routine like watching always the same TV program at the same time.

There is no single known cause of autism although recent research has provided strong evidence for a genetic factor underlying many instances. In other instances, however, there is evidence of brain damage occurring before, during or soon after birth.

There is no known cure for autism as yet, but a combination of early intervention, special education, speech therapy, occupational therapy, behav-


206

ioural interventions and structured support has been found to help develop important skills, minimize behaviour problems and thereby enable autistic individuals to achieve as much independence as possible.

Experience has also shown early diagnosis and intervention to be critical for the development of the individual in question and essential to ensure ac-cess to appropriate services and professional support.

THE EUROLAND PROJECT

The Euroland project was funded by a Marie Curie scholarship granted to the second author of this paper by the European Community. The purpose of the project was to construct a three-dimensional (i.e., 3D) virtual world called Euroland.

The community of Euroland consisted of 42 pupils and 10 teachers from 4 Italian and 3 Dutch classrooms. The seven classrooms had a shared interest in the implementation of new educational technologies and were very moti-vated to work together.

The schools involved also had sufficient computer facilities. In addition to the pupils and teachers, 9 researchers — both Italian and Dutch — helped realize the project and performed various tasks (e.g., on-line tutoring, class-room observation, on-line and classroom coordination of the project, techni-cal support, pedagogical support). Only a small group of pupils from each classroom or so-called ‘citizens’ connected regularly to Euroland and thereby constituted a bridge between classmates and the virtual community of Eu-roland.

The 3D virtual world was created via the integration of two computer software programs: Active Worlds (AW) and Web Knowledge Forum (WKF). Using AW, it is possible to create a virtual world with houses, roads, trees and other objects; it is also possible for users to chat synchronously with other connected users (Bricken, 1991).

WKF is a discussion forum developed by Scardamalia and Bereiter (1994). At the centre of WKF is a communal database, which can be filled with con-tributions or ‘notes.’

All of the notes are saved in the database and thus available to all partici-


207

pants. The WKF also provides space for the sharing of documents and files, or a so-called Shared Directory.

The virtual space created with the combination of the two aforementioned pieces of software provided numerous opportunities for communication and collaborative learning (Ligorio, 2001). The participants in the project indeed established a complex virtual community and gradually occupied the virtual space which was completely empty at the outset of the project (Talamo, Zuc-chermaglio, Ligorio, 2001).

In a virtual world, pupils undertake several activities to further plan, con-struct and evaluate the world while it is under construction. Pupils meet in a virtual space and encounter peers and adults; they can develop new social and communication skills; they can develop new practices; they can learn traditional school content along with information specific to the tools being used; they can form virtual communities; they can build new knowledge as a consequence of the construction and decoration of virtual houses; and they can acquire new cultural values.

In fact, such pupils shape the virtual environment they populate, both physically and culturally (Ligorio, van der Meijden, submitted). The Euroland community was very active during the eight months of its existence from October 1999 to May 2000.

And this was true for both interactions inside each classroom and within the virtual world. Interactions occurred between pupils, teachers and research-ers mostly in English, which was a second language for all of the participants. The communication skills specific to the various communication formats were found to improve during the project, which means that the participants became progressively more expert at chatting, discussing on the Forum, and using 3D images (Ligorio, 2001).

Pupils were found to work on and discover the educational value of topics related to the regular curriculum (e.g., history, music, art) and topics of their own choosing (e.g., food, sports, music, travel).

Eight cultural houses were constructed and decorated in Euroland. An-other fourteen virtual houses were constructed as purely social or individual spaces or for the exchange and discussion of information.


208

FIGURE 1. Euroland went from an empty world to a world full of cultural housesand original activities

FIGURE 2. List of virtual houses built in Euroland


209

DATA COLLECTION AND ANALYSIS

An enormous amount of data was collected: 80 hours of chat, 8 virtual houses with their cultural content and decorations, 162 notes posted on the WKF and 102 files uploaded to the Shared Directory of the WKF. The data was directly recorded by the software in the format of chat-logs and material stored on the forum.

The researchers also took screen pictures to keep track of the changes occurring in the virtual houses. The data was analyzed in a mainly qualita-tive manner. The ethnographic method (Duranti,& Goodwin, 1992) allowed analysis of the project over time. And via a longitudinal description of the cyberspace occupied by the Eurolanders, it was possible to track the transfor-mation of Euroland from an empty space to a world full of cultural houses and original activities.

For purposes of the present study, a discourse analysis was undertaken to single out the most relevant topics (Sacks, 1992; Schegloff, 1992) occurring during Euroland communication.

The analysed discussions were selected from the 80 hours of chat col-lected, based on the presence of Ivo, the protagonist of our case study. The other types of data (e.g., materials and notes posted on WKF, pictures of 3D objects, like the exhibition on representations of Venus, made by Ivo) were used to support the interpretation of the chat data.

IVO ON-LINE

To illustrate how an actual classroom culture can change as a result of involvement in a virtual community, the development of Ivo — a secondary school pupil diagnosed with a form of autism — will be considered here in detail.

Ever since one of Ivo’s science teachers showed him the computer pro-gram Active Worlds during an early stage of his school career, Ivo showed a keen interest in computers.

When Euroland was started, Ivo was already an AW expert and quickly volunteered to participate in the project. Neither the on-line tutors nor the


210

remainder of the community were informed of the fact that Ivo was affected by autism.

The data collected and particularly the chats show Ivo to clearly prefer interaction with (mostly adult) on-line tutors and to limit his interactions with peers and other adults (i.e., teachers from other schools and researchers).

As can be seen from the following excerpt of a chat session between Ivo and another pupil from Ivo’s school participating in Euroland, Lars, and Clarence and Bea, tutors from Italy, Ivo freely requests information about building and the materials to use.

Excerpt 1: January 12, 2000NICKNAME LOG FILE OF THE CHAT

EDUHML1 Ivo: We could make a house and in that house we could make models of typical Dutch & Italian houses to compare them

Clarence:2 yes, I think it’s a great ideaEDUHML Lars: SUPAMANClarence: Lol 3

EDUHML Ivo: But who makes the models??Clarence: we can find pics in on the netEDUHML Ivo: I think we’d best make them in Calligary Truespace2

EDUHML Ivo: No, not pics, MODELS EDUHML Ivo: :) 4

Clarence: Ivo....you and Lars and all in EDUHML are very expertClarence: and you are able to build your own models

EDUHML Ivo: How about an Italian from the late XIX century, the 50’s & less than 10 year old build-ings?

Clarence: others can hardly build with standard building objects !EDUHML Lars: they don’t?EDUHML Lars: we can teach themEDUHML Ivo: At 1st we wanted to make our own objects but that is not completely necessaryClarence: some can hardly build, others simply can’tClarence: we are here to teach them, indeedClarence: but Truespace is way too farEDUHML Ivo: ?Clarence: too difficult I meanBea: Yea my concern about your idea, Ivo/Lars, is that you will go much further ...EDUHML Ivo: We’ll make themClarence: I agree with your project, but try to use standard objectsEDUHML Ivo: than...Clarence: we have in AW universeClarence: perfect imitation of existing Italian palaces


211

Clarence: like Coliseum, palazzo della signoria in Firenze and moreEDUHML Ivo: AhaClarence: made using standard objects EDUHML Ivo: I C 5

EDUHML Ivo: OK

Bea: Ivo Lars, I would prefer if you could share your ideas with the Italians and discuss with them and doing it together ...

Clarence: Ivo and Lars.....I think you are very good in at buildingBea: instead of having your own place ...Clarence: but try to think of other’s abilitiesBea: do you understand what I mean?EDUHML Ivo: NoEDUHML Ivo: :0

Bea: I mean: if you have your own building you will go much further comparing to the other in terms of AW expertise ...

Bea: I think it would be better if you could work more closely with the others ...Bea: sharing idea, and building together...EDUHML Ivo: Ehhh? I don’t follow it anymore.Bea: helping the others in advancing ...

1. This stands for the name of the school.2. Calligary Truespace is a 3D building program.3. Lol means ‘laughing out loud’ in chat talk.4. This is an emoticon which simulates a smile. 5. IC means ‘I see’ in chat talk

Ivo and Lars indeed have become very skilled builders with the support of the Dutch science teacher who stimulated their involvement in Active Worlds and Euroland. As the excerpt shows, they found the standard objects to be of little interest and felt the need to create new objects using a more sophisticated program, Calligary Truespace. The 3D objects created using such software can be transferred into Euroland.

The tutors Clarence and Bea are concerned about the gap between these Dutch students and Italian pupils, with the latter hardly able to build at all. The tutors therefore try to place the skills of the two Dutch pupils at the service of the entire community and use such terms as ‘collaboration’ and ‘the other.’

Every time the tutors turn the conversation to these topics, Ivo can be seen to react in a rather strange manner and simply not understand the pro-posal. In other chat sessions, Ivo just ignores any remarks regarding collabo-ration or leaves with one or the other excuse.

In the excerpt which follows, the tutor again tries to convince Ivo to be-


212

come a technical tutor for other less skilled Euroland citizens. In order to do this, however, it is necessary to establish contact with the other pupils by chatting and interacting with them; the initial contact can be about general topics, in this case sports, but Ivo states that he does not know anything of sports and he leaves the chat room.

Excerpt 2: February 9, 2000NICKNAME LOG FILE OF THE CHAT TRANSLATION IN ENGLISH

Bea: hello Ivo :))EDUHML Ivo: CIAO BE(LL)A

Bea: there is maurizio .. you may talk to him about it ...

EDUHML Ivo: I’ll do that

Bea: so maurizi do you have questions for Ivo ??

Bea: Ivo come closer ...

Bea: (to Maurizio) dai approfittatene che mica capitano tutti i giorni gli olandesi in rete :) 6

(to Maurizio) you better profit, Dutch are not con-nected every day

Maurizio: ok ci pensiamo noi Ok let us do it

Maurizio: Ivo do you like the Dutch sport house we are building?

Maurizio: ma Ivo parla inglese? But does Ivo speak English?Bea: (to Maurizio) Certo (to Maurizio) Of course…EDUHML Ivo: But I don’t know anything about sportsEDUHML Ivo: I’m sorry, I have to go

6. Italics indicate the use of the ‘whisper’ function or, in other words, two participants having a private con-versation (which happened to be in Italian here). The utterances typed using the whisper function are not visible to other participants which meant that Ivo could not read this interaction.

The tutor Bea is trying to get Ivo and Maurizio, who is one of the Italian pupils, to talk to each other. This does not appear to be easy for Ivo because the topic of the conversation is something which he knows nothing about and it does not concern how to build a 3D object, which he knows a lot about.

The fact that he does not communicate, despite encouragement from the tutor, is noticed by the Italian pupil and mistakenly interpreted as a sign that Ivo’s English skills are not sufficient to communicate with others.

Most of the time Ivo works alone, and the result is a great Venus exhibi-tion (see below).


213

The tutor then suggests that Ivo organizes a guided tour for peers and other visitors. Ivo agrees and prepares a number of questions which he then posts on the forum (e.g., ‘What do you think is the oldest painting?’ or ‘If you look at the pictures, do you think the concept of beauty has changed over time?’).

When he notices that the community appreciates his work and receives numerous reactions to his questions, Ivo’s style of communication appears to change: he becomes more open and his interactions with other pupils be-come longer and more complex.

FIGURE 2. Ivo’s Venus exhibition

To illustrate how Ivo’s role has changed within Euroland, a short excerpt of a chat session in which Clarence — one of the Italian tutors — tries to explain some technical features to two Dutch newcomers is presented.

As can be seen, the girls have difficulties with the explanation being pro-vided by the tutor and Ivo gradually takes over.


214

Excerpt 3: February 21, 2000NICKNAME LOG FILE OF THE CHAT TRANSLATION IN ENGLISH

Clarence: open the ‘Teleport’ menu in the upper left side of the program

Iris: we moeten hem volgen, Anne We must follow him, AnneClarence: click on ‘To...’Clarence: a little window will openClarence: write down these coordinates:Anne: ivo bedoel ik I mean IvoClarence: 3s 6eClarence: this is a way to be teleported where you want,

by writing the coordinatesClarence: now please tryIris: ik snap het niet jij anne I don’t understand, do you, Anne?EDUHML Ivo: Willen jullie uitleg? Do you want an some explanation?Anne: ik ook niet heeft hij het tegen ons? Me neither, is he talking to us?Anne: ja graag Yes, pleaseEDUHML Ivo: Jullie willen naar de coordinaten 3s 6w? You want to go to the coordinates 3 South

and 6 West, right?Anne: okEDUHML Ivo: Je drukt op de Alt-toets You press the Alt buttonEDUHML Ivo: Dan T Then ‘T’EDUHML Ivo: Nog een keer T Then ‘T’ againEDUHML Ivo: Dan zie je een veld met ‘Euroland’ You’ll then see a field with ‘Euroland’

The preceding interaction concerns a technical detail — namely, how to move around in Euroland — which happens to be one of Ivo’s favourite top-ics. Furthermore, Ivo was able to chat with the girls in Dutch. Nonetheless, from this chat on, Ivo interacts much more often with other pupils and the topics of the interactions are less frequently about technical details over time. At the end of the project, Ivo’s expertise is generally recognized by the entire community; he is frequently consulted; and his advice is clearly appreciated and considered valuable.

CHANGING CLASSROOM CULTURE

Ivo’s behaviour in the classroom is found to change as well. When Ivo entered secondary school, he had no contact with peers; he did not make eye contact with either teachers or peers; and he was totally involved with his own person. In the classroom, he mostly sat with his head in his hands and


215

arms on the desk, and he was often alone. Slowly his expertise as an AW and Truespace builder was recognized within the classroom and he was asked for advice. Parallel to the changes on-line, Ivo’s behaviour and interactions in the classroom also changed. He was frequently invited to present the Euroland project to other classrooms or to an external audience. Ivo became more and more capable of orally presenting the project in public and demonstrating the virtual world. While he sometimes got a little confused and he was not always able to remember the line of his presentation, his self-confidence could be seen to increase gradually and constantly.

AFTER EUROLAND

After completion of the Euroland project, the school which Ivo attended was involved in two more AW projects i n collaboration with the Municipal Museum in The Hague and the Department of Educational Sciences at the University of Nijmegen. Both of the projects were about art (Van der Mei-jden, Janssen, Ligorio, 2001). Ivo decorated the two 3D-worlds with objects created in Truespace and sometimes did this at the request of pupils who did not know how to use the AW program. In such a manner, Ivo not only developed his own technical skills but also helped other pupils develop their skills.

After his graduation, Ivo has been accepted to the High Vocational School of Arts with a specialization in 3D Design. For this pupil with autism, the computer, participation in 3D design and building activities and collaboration with others clearly contributed to his cognitive and psychosocial develop-ment. While individual computer use carries the risk of isolating pupils, the use of the computer for collaborative purposes — as done here — can instil feelings of belonging and thereby promote the process of socialization.

In closing, we would like to report that pupils and teachers evaluated the Euroland project very positively. The reactions following completion of the project were very positive with regard to not only the classroom culture but also individual progress and particularly the development of pupils with learning problems.


216

REFERENCES

American Psychiatric Association (1987), Diagnostic and statistical manual of mental disorders, Third edition, revised, Washington, DC: Author.

Baron-Cohen, S., Tager-Flusberg, H., Cohen, D.J. (2000) (Eds.), Understanding other minds: perspectives from developmental cognitive neuroscience, Oxford: OUP.

Bordia, P. (1997), “Face-to-face versus computer-mediated communication: A synthesis of the experimental literature”, in Journal of Business Communi-cation, 34(1), 99-120.

Bricken, M. (1991), “Virtual reality learning environments: potential and chal-lenges”, in Computer Graphics, (25)3, 178-84.

Caravita, S., Ligorio, M.B. (2003) (Eds.), Apprendimento collaborativo. Dal gruppo alla rete. [Collaborative learning. From the group to the net], Rome: Edizioni Carlo Amore http://www.carloamore.it/

Dillenbourg, P. (1999), “Introduction: What do you mean by ‘collaborative learning’?”, in P. Dillenbourg (Ed.), Collaborative learning. Cognitive and com-putational approaches, Oxford: Pergamon, pp. 1-19.

diSessa, A. (2000), Changing Minds: Computers, Learning and Literacy, Cambridge: MIT Press.

Duranti, A., Goodwin, C. (1992) (Ed.), Rethinking Context: Language as an Inter-active Phenomenon, Cambridge: Cambridge University Press.

Goldenberg, E.P., Russell, S. J., Carter, C. J. (1984), Computers, education and special needs, Massachusetts: Addison-Wesley Publishing Company.

Johnson, R. T., Johnson, D. W. (1986), “Action research: Cooperative learning in the science classroom”, in Science and Children, 24, 31-32.

Ligorio, M. B. (2001), “Integrating different formats of communication: syn-chronous versus asynchronous and text-based versus visual”, in Computers & Education, Vol. 37/2, 103-125.

Sacks, H. (1992), “Lectures on Conversation”, 2vols., Gail Jefferson (Ed), in Human Studies, 12 (3/4), 183-404.

Scardamalia, M, Bereiter, C. (1994), “Computer support for knowledge-build-ing communities”, in The Journal of the Learning Sciences, 3, 265-283.

Schegloff, E.A. (1992), “On talk and its institutional occasions”, in P. Drew, J. Heritage, (Eds.), Talk at Work. Interaction in institutional settings, Cambridge: Cambridge University Press, pp. 101-134.


217

Slavin, R. E. (1990), Cooperative Learning: Theory, Research and Practice, Needham Heights: Allyn and Bacon.

Talamo, A., Nimivierta, M. (2000), “An overview of motivational measures”, in H. Van der Meijden, F. De Jong, R.J. Simons (Eds.), Computer supported collaborative learning networks in primary and secondary education, Final Report Project EC 2017, Nijmegen, The Netherlands: University of Nijmegen, pp. 81-93.

Talamo, A., Zucchermaglio, C., Ligorio, M.B. (2001), “Communities’ Devel-opment in CVEs and sustaining functions of on-line tutorship”, in G. Riva and C. Galimberti (Eds.), CYBERPSYCHOLOGY: Mind, cognition and society in the Internet Age, IOS Amsterdam.

Van der Meijden, H., Janssen, J., Ligorio, M.B. (2002), “Edgar Degas: Re-constructing his art in a three-dimensional virtual world”, Proceedings of the Conference ‘Dealing with diversity’ of the International Society for Cultural Research and Activity Theory, Amsterdam, June 18-22, 2002, pp.71-72.

Van der Meijden, H., Veenman S. (in press), “Face-to-face versus computer-mediated communication in a primary school setting”, Computers in Human Behavior

Wetherby, A.M., Prizant, B.M. (2000) (Eds), Autism spectrum disorders: A trans-actional developmental perspective, Baltimore: Paul Brookes.

ABSTRACT

Numerous studies have shown collaborative learning to enhance not only pupil achievement but also social development (Dillenbourg, 1999; Johnson, Johnson, 1986; Slavin, 1990). The introduction of electronic networks and virtual communities in schools can create new cultures and thereby change the classroom routine. In this paper, we will describe the development of Euroland which is a joint project between Italy and The Netherlands. Spe-cial attention will be paid to one of the Dutch pupils participating in the project for two reasons: first, his high level of computer skills; second, his learning and psychosocial problems as a result of autism. It is then shown that the collaboration and communication occurring during the creation of a three-dimensional virtual world can change not only the classroom culture


218

but also individual performance in such a manner that children with learning problems are better integrated into both the school community and a wider community.

Received October 13, 2003; accepted February 20, 2004


219

RUNNING HEAD: SCHOOL SETTINGSCHOOL CLIMATE AND WELL BEING IN

EARLY ADOLESCENCE: A COMPREHENSIVE MODEL

Alessio Vieno*1, Massimo Santinello*, Elena Galbiati*, Massimo Mirandola**

INTRODUCTION

School-related stress influences to the development and maintenance of students’ self-esteem, self-efficacy, health, and well-being (Samdal et al., 1998). The influence of school climate on students goes beyond academic educa-tion, but little interest has been shown thus far in investigating the contribu-tion of school climate on students’ well-being and lifestyles (Allison et al., 1997). The aim of the present work is to present and test a theoretical model analysing various factors involved in school settings and how they influence adolescents’ health and well-being.

Epidemiological studies have highlighted how between 12% to 30% of school-age children experience at least a moderate degree of behavioural, social, and emotional problems (Hoagwood et al., 1996; Hosman, 2000; In-stitute of Medicine, 1994). Given that some of the cited problems can be exacerbated during the transitional period of middle school (Kazdin, 1993), it is particularly crucial to study the effects of school functioning (achievement, integration, and needs satisfaction) throughout this developmental phase. In fact, during early adolescence, students experience discontinuity in their school life, which results more difficult because of the passage from middle school to high school (Galbiati, Santinello, 2002). In addition to the difficul-

*Department of Developmental and Social Psychology. University of Padova.**Regional Centre for Health Promotion, Veneto Region.1Address correspondence to: Alessio Vieno, Dipartimento di Psicologia dello Sviluppo e della Socializzazione LIR-IPAC – Via Belzoni 80, 35131 Padova – e-mail: [email protected]

Alessio Vieno et al.

220

ties this transitional phase entails, Italian 8th graders must make decisive and binding choices concerning their future scholastic and career paths.

School-related experience can therefore represent either a risk factor or a resource for early adolescent and adolescent development. Students who do not like school are more exposed to the risk of school failure, which can have a long-term impact in terms of being at risk for unhealthy behaviour, psychosomatic problems, and a poorer quality of life (Aunola et al., 2000; Butters, 2002; Hurrelmann et al., 1995; Jessor, 1991; Nutbeam et al., 1993; Samdal et al., 1998).

The relation between school climate and scholastic achievement has been well-documented, with the finding that students with a good perceived class climate are more academically motivated than less-satisfied students (Battis-tich et al., 2000; Fraser, 1994; Mortimore, 1995; Resnick et al., 1993; Roeser et al., 1998; Roeser et al., 2001).

Student satisfaction, achievement, and school-related stress have been found to depend on individual characteristics (e.g., sociability and the positive values that are associated with academic achievement, Couvington, Beery, 1976) as well as on the social characteristics of families, peers, and school settings (Perry et al., 1993; Thorsheim, Wold, 2001).

Other studies reported in the literature suggest that a positive school cli-mate and high academic achievement are associated with a good relation-ships with teachers (Bost et al., 1998; Cohen, Syme, 1985; Harter, 1996; Step-toe, 1991; Wentzel, 1998), with parents (Connell et al., 1994; Gore, Aseltine, 1995), and with peers (DuBois et al., 2002; Tani, 2000; Wasserstein, La Greca, 1996).

The focus in the above-cited studies was on how individuals perceive and experience the characteristics of their surrounding social setting. It is in this sense, then, that peer, parent, and teacher support relationships as well as appropriate expectations (especially teachers’ and parents’ expectations) can contribute to academic achievement.

These factors can also help students perceive a positive school climate and reduce school-related stress (Galbiati, Santinello, 2002). All of these charac-terizing elements of school settings have an impact, in turn, on substance abuse (Bonino, Cattelino, 2000; Nutbean et al., 1993; Perry et al., 1993), per-ceived individual health (Knishkowy et al., 1995; Samdal et al., 1998; Torsheim,

School climate and well being in early adolescence

221

Wold, 2001), and student quality of life (Argyle, Martin, 1991; Cattelino, 2000; Graham et al., 1997; Nutbean et al., 1993).

In particular, it has been shown how students who adapt better to their school setting and with higher academic achievement, perceive themselves as healthier, say they are happier, and are less likely to adopt risky health behav-iour, such as substance abuse.

The theoretical model Based on part of the above-described experimental evidence, Samdal,

Wold, and Torsheim (2000) proposed a theoretical model (Fig. 1) that re-lates the various components of school settings to early adolescent well-be-ing and health. In particular, the first group of variables (concerning aspects of perceived social support and the expectations of the three main figures in adolescents’ lives — their teachers, parents, and classmates) are considered predictors of students’ relations with their school setting and its various com-ponents, e.g., satisfaction, academic achievement, and school-related stress. In turn, this second set of variables is considered to be related to student health and quality of life as well as a protective factor against substance use (alcohol and tobacco).

FIGURE 1. The theoretical model (source Samdal et al., 2000)

The theoretical model has been only partially tested (Bru et al., 2002; Torsheim, Wold, 2001), by separately examining either some of the model’s

Teachersupport

Parental support

Classmatesupport

Teacher/ParentSupport

Satisfactionwith school

Academicachievement

School-relatedstress

Riskbehavior

Perceivedhealth

Qualityof life


222

school climate components or diverse health outcomes. The present study therefore proposes to test the theoretical model by

comprehensively examining various school climate components in relation to different aspects of well-being.

Therefore, this study specifically aims at verifying the following hypoth-eses concerning early adolescent students:

• good relationships with teachers, parents, and peers as well as perceived support from these figures are associated with school satisfaction, higher academic achievement, and less school-related stress;

• conversely, perceived teachers’ and parents’ academic demands are negatively associated with school satisfaction and achievement, and are positively linked to school-related stress;

• satisfaction with school and a good perception of one’s own academic achievement are positively linked to quality of life and perceived health, whereas they are negatively associated with risk behaviour;

• perceived school-related stress is positively related to risk behaviour and negatively associated with health and well-being indicators.

THE STUDY

The data presented in this study are part of the “Health Behaviour in School-aged Children” (HBSC) project, a trans-national study carried out in collaboration with the European office of the World Health Organisation. HBSC was developed to investigate and improve the understanding of early adolescent and adolescent health behaviour (Wold et al., 1994).

The international group coordinating the research project develops and tests the quality of questionnaire items, adapting them to the various cultural contexts of the nations involved in the project (WHO, 1998).

Participating nations are required to follow the sampling criteria set forth in the Research Protocol, registered at the World Health Organisation Re-gional Office, and administer the questionnaire developed by the interna-tional group.


223

Participants The questionnaire was filled out by a total of 1,347 early adolescent, 8th

grade students (48.6% girls) from 82 middle schools in the northern Italian region of Veneto. Participants’ mean age was 12.99 years.

ProcedureTwo-stage cluster sampling (Thompson, 2002) was used to select the sample.

The questionnaire was administered to all students in the selected classes. Schools were selected (82 from 351) by stratified sampling based on geo-graphical area (in terms of each school’s local health authority). This design ensured that our sample accurately represented the referent population (early adolescent inhabitants of the Veneto region).

The questionnaire was anonymous. The students’ teachers administered the questionnaires, and students took from 45 minutes to one hour to fill in the questionnaires.

VariablesTeacher support. Four items were used to assess teacher support (Torsheim et

al., 2000). Students were asked to express their degree of agreement with the following statements about their teachers: (1) “I am encouraged to express my own views in my class(es)”, (2) “Our teachers treat us fairly”, (3) “When I need extra help, I can get it”, (4) “My teachers are interested in me as a person”.

The answers to these items were distributed on a 5-point Likert scale (ranging from “strongly disagree” to “strongly agree”). The internal consis-tency analysis for this sum score yielded a Cronbach’s α of .75.

Parental support. Three items were used to evaluate parental support (Tor-sheim et al., 2000). The items were as follows: (1) “If I have a problem at school, my parents are ready to help”, (2) “My parents are willing to come to school to talk to teachers”, (3) “My parents encourage me to do well in school”.

The answers were distributed along a 5-point Likert scale (ranging from “never” to “always”). The internal consistency analysis for this sum score


224

yielded a Cronbach’s α of .68.Classmate support. Four items were used to assess classmate support (Tor-

sheim et al., 2000). The items were as follows: (1) “The students in my class enjoy being together”, (2) “My classmates accept me as I am”, (3) “Most of the students in my class(es) are kind and helpful”, (4) “How often does it happen that other students don’t want to spend time with you at school and you end up being alone?”.

The answers were distributed along a 5-point Likert scale (ranging from “never” to “always”). The last item was re-coded to make it consistent with the rest of the scale. The internal consistency analysis for this sum score yielded a Cronbach’s α of 66.

Teacher and parental academic expectations. Two items examined if students perceived parent and teacher expectations as unrealistic. Students were asked to indicate their degree of agreement with the following statements concern-ing their teachers: (1) “My parents expect too much of me at school“, (2) “My teachers espect too much of me at school”. The answers to these items were distributed on a 5-point Likert scale (ranging from “strongly disagree” to “strongly agree”). The correlation (Spearmans) between these two questions was r = .49 (p < .001).

Satisfaction with school. Three items from Huebner et al.’s (1998) scale were used to assess school satisfaction. The items were as follows: (1) “How do you feel about school at present?” (on a 4-point Likert scale ranging from “I don’t like it at all”, to “I like it a lot”), (2) “Our school is a nice place to be” (on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree”), (3) “How often do you think going to school is boring?” (on a 5-point Likert scale ranging from “never” to “very often”). The last item was re-coded to make it consistent with the rest of the scale. The internal consistency analysis for this sum score yielded a Cronbach’s α of .77.

Academic achievement. One item was used to assess students’ academic achievement (Samdal et al., 1998): “In your opinion, what does your class teacher(s) think about your school performance compared to your class-mates’?” (on a 4-point Likert school ranging from “below average” to “very good”).

School-related stress. The item “How pressured do you feel by the schoolwork you have to do?” was used to assess the degree of academic stress students


225

perceive (on a 5-point Likert scale ranging from “not at all” to “a lot”).Risk behaviour. Four items were used to assess risk behaviour (Dawson,

1998): (1) “How often do you smoke tobacco at present?” (on a 4-point Lik-ert scale ranging from “I don’t smoke” to “every day”), (2) “How often do you drink beer? (3) “How often do you drink wine?”, and (4) “How often do you drink spirits?”. The answers to the 3 questions concerning alcohol use were distributed on a 5-point Likert scale (ranging from “never” to “every day”). The internal consistency analysis for this sum score yielded a Cron-bach’s α of .70.

Perceived health. Nine items were used to assess students’ perceived health. The first item (Idler, Benyamini, 1997) concerned general health: (1) “How healthy do you think you are?”(on a 3-point Likert scale ranging from “not very healthy” to “very healthy”). The remaining 8 items were taken from the “HBSC symptom checklist” (Torsheim, Wold, 2001). Students were asked to indicate “In the last six months: How often have you had the following?: (1) “headache”, (2) “stomach-ache”, (3) “back-ache”, (4) “feeling low”, (5) “ir-ritability or bad temper”, (6) “feeling nervous”, (7) “difficulties in getting to sleep”, (8) “feeling dizzy”. The answers to the 8 items were distributed along a 5-point Likert scale (ranging from “about every day” to “rarely or never”). The internal consistency analysis for this sum score yielded a Cronbach’s α of .76.

Quality of life. The item “In general, how do you feel about your life at present?” was used to assess the degree of students’ perceived quality of life; students responded on a 4-point Likert scale (ranging from “I don’t feel happy at all” to “I am very happy”).

Statistical analysis The first analyses were conducted with the SPSS (8.0) programme. Cron-

bach’s Alpha was used to assess internal consistency of the scales; for degree of correlation among items, the bi-variate correlation coefficient was used.

Structural equation models method (Jöreskog, Van Trillo, 1973), imple-mented by the program LISREL (8.50), was used to verify the adaptability of our data to the model proposed by Samdal et al. (2000) and to avoid limiting our attention exclusively to cause-and-effect relations among independent


226

and dependent variables. This technique provides a more complete picture of a given situation, by additionally examining causal mechanisms among independent variables. Hence, the method allows us to examine not only the multiplicity of causes acting on a dependent variable, but connections among various causes as well (Corbetta, 1992).

The model represents the first four variables (teacher support, parental support, classmate support, and demands) as exogenous variables, and the other variables are represented as endogenous variables.

Whereas exogenous variables act exclusively as independent variables, en-dogenous variables can either be dependent or independent variables. The γ values represent the parameters expressing the strength of connections among endogenous and exogenous variables, while the β values represent links among exogenous variables; the φ values represent coefficients of cor-relation between variables.

We considered the following indices as indicators of the model’s overall goodness of fit: 1) Chi square (χ²); 2) the Goodness of Fit Index (GFI), the Adjusted Goodness of Fit Index (AGFI), and the Comparative Fit Index (CFI)--with values ranging from 0 (a poor fit) and 1 (a perfect fit); and 3) the Root Mean Squared Error of Approximation (RMSEA), considered accept-able when it is lower than .08 (Cudeck, Browne, 1992).

ResultsTable 1 shows the descriptive statistics and correlational scores among the

variables included in the model. It is interesting to note how, although nearly all the correlation scores were significant, almost all of them were rather low (< .50).

The relation between perceived health and quality of life represented the highest correlation (.37), confirming how the physical and psychological components of health are closely related.

The correlation between teacher support and school satisfaction (.35) rep-resents a further interesting result, confirming a common finding in the lit-erature (Cohen, Syme, 1985)--i.e., that the social support of teachers is a key element in the relations students establish with their school setting, in terms of satisfaction and achievement.


227

Item 1 2 3 4 5 6 7 8 9 M DS

1. Teacher support - 2.91 .83

2. Parental support .28** - 2.05 .84

3. Classmate support .17** .18** - 2.55 .78

4. Demands -.10** -.06* -07** - 3.23 .88

5. Satisfaction .35** .22** .12** -.19** - 3.27 .92

6. Achievement .33** .15** .07** -.18** .33** - 2.53 .86

7. School-related stress -.13** -.08** -.12** .18** -.26** -.04 - 2.57 .89

8. Risk behaviour -.15** -.16** .01 .13** -.26** -.17** .02 - 2.80 .43

9. Perceived health .14** .22** .20** -.10** .16** .07** -.09** -.16** - 2.63 .51

10. Quality of life .18** .24** .24** -.10** .21** .16** -.09** -.12** .37** 1.97 .76

** p < .01; * p < .05.

TABLE 1. Means, standard deviations, and correlation coefficients among the model’s measures

To test for differences between the covariance matrices for males and fe-males, procedures proposed by Byrne (1989) were used.

The analysis yielded statistically significant results (χ² = 525.02, p = .001), and we therefore proceeded by testing the model separately for boys and girls. Figure 2 represents the empirical verification of the theoretical model on the boys’ sample.

Non-significant parameters were excluded to limit the number of param-eters in order to make the results easier to interpret. Parameters were elimi-nated one at a time, with a subsequently new calculation conducted each time on the model (Corbetta, 1992).

The calculated structural β and γ coefficients were therefore significantly different from zero and consequently, the yielded relations were all statisti-cally significant. The general fit indexes obtained for the boys’ model were as follows: χ2(22) = 102.19, p = .01, GFI = .97, AGFI = .93, CFI = .86, RMSEA = .07, indicating an acceptable degree of fit for our data to the proposed theoretical model.


228

FIGURE 2. Model showing standardized path coefficients for boys

The R2 values, which were calculated for each endogenous variable on the reduced form (Jöreskog, 2000), were as follows: R2

y1 = .13, R2y2 = .04, R2

y3 = .05, R2

y4 = .06, R2y5 = .03, R2

y6 = .07. The final model thus accounted for the following percentages of variance: 13% of satisfaction with school, 4% of aca-demic achievement, 5% of school-related stress, 6% of risk behaviour, 3% of perceived health, and 7% of quality of life.

In accordance with the results of our descriptive analysis, school satisfac-tion was most strongly linked to teacher support (γ = .32). Parental support also had a significant influence on this variable (γ = .08), although the ef-fect was much lower than that of teacher support. Conversely, teachers’ and parents’ demands had a negative influence on school satisfaction. Teacher support was also linked to academic achievement, and this variable was also related to parents’ and teachers’ expectations. This last variable was also re-lated to perceived school-related stress, which was also linked to classmate support.

Moreover, the model shows how school satisfaction was linked to all health outcomes, and how academic achievement was the first predictor (β = .14) of quality of life. Contrary to what hypothesized, school-related stress was not associated to the health outcomes.

Figure 3 represents empirical verification of the model on the girls’ sam-ple. Once again, non-significant parameters were eliminated one at a time, with subsequently new calculations being made on the model (Corbetta,

Teachersupport

Parental support

Classmatesupport



Academicachievement


Riskbehavior

Perceivedhealth

Qualityof life

χ2 = 102.19, p = .01, DF = 22, GFI = .97, AGFI = .93, CFI = .86, RMSEA = .07

.32

.15

.08

-.16

-.15-.11

.17

-.19

.11

.12

.07

.13

.14

.12

.16

-.02

.15


229

1992). The model’s calculated structural β and γ coefficients were therefore all significantly different from zero, and thus, the relations emerging were all statistically significant.

General fit indexes obtained for the model were as follows: χ2(19) = 113.41,

p = .01, GFI = .97, AGFI = .91, CFI = .88, RMSEA = .08, once more reveal-ing an acceptable degree of fit for our data to the proposed model.

FIGURE 3. Model showing standardized path coefficients for girls

The R2 values, calculated for each endogenous variable on the reduced form (Jöreskog 2000), were as follows: R2

y1 = .18, R2y2 = .10, R2

y3 = .04, R2y4

= .08, R2y5 = .06, R2

y6 = .07. The final model therefore accounted for the fol-lowing percentages of variance: 18% of satisfaction with school, 10% of academic achievement, 4% of school-related stress, 8% of risk behaviour, 6% of perceived health, and 7% of quality of life.

As with boys, girls’ school satisfaction was most strongly linked to teacher support (γ= .29). Parental (γ= .17) and classmate support (γ= .10) also show a significant, though lesser, effect on school satisfaction. Teachers’ and parents’ demands had a negative influence on this variable. Essentially the same trend was found for academic achievement, i.e., the positive influence of parental and teacher support, vs. demands, which had a negative effect. As with boys, we found a positive relation between teacher expectations and school-related stress, but whereas classmate support was a protective factor for boys, it was teacher support that had this effect on girls.

Teachersupport

Parental support

Classmatesupport



Academicachievement


Riskbehavior

Perceivedhealth

Qualityof life

χ2 = 113.41, p = .01, DF = 19, GFI = .97, AGFI = .91, CFI = .88, RMSEA = .08

.29

.16

.17

.10

-.13-.16

.12

-.10

.10

.14

.23

.06

.12

-.08

.23

.11

-.06

-.14

.14

-.09

.16

.10

-.05

.15

-.03

-.03


230

The figure 3 shows how school satisfaction was also linked to girls’ health outcomes. Yet, girls seemed to turn to risk behaviour, unlike their male class-mates, who showed a relation between school achievement and quality of life. Lastly, whereas boys apparently did not suffer unduly from school-related stress, in girls, this variable was linked to perceived health.

DISCUSSION

The results of the study essentially confirm Samdal et al.’s (2000) theoreti-cal model. According to this model, during the early adolescent transitional period, school climate affects not only students’ motivation and school sat-isfaction, but their lifestyles, health, and quality of life as well. Furthermore, although the role of teachers takes on different nuances for boys and girls, our results underscore the meaning and importance of this figure in the bio-psychosocial well-being of early adolescents. Therefore, in agreement with previous research (Bost et al., 1998; Cohen, Syme, 1985; Harter, 1996; Sam-dal et al., 1998; Steptoe, 1991; Wentzel, 1998), teacher support is essential to creating a school climate characterized by interaction and teamwork aimed at fostering feelings in students of being listened to and understood. Moreover, these conditions are related to students’ well-being and health.

The causal relations model tested on our males’ subsample also highlights how different support figures can take on different roles. If teacher sup-port is crucial to boy’s school satisfaction and achievement, boys’ classmates also play a crucial role in helping them cope with school-related stress (Was-serstein, La Greca, 1996), which is related to parental and teacher demands (Perry et al., 1993; Samdal et al., 2000; Steptoe, 1991).

For boys, parental figures were related only marginally to school satisfac-tion (Connel et al., 1994; Gore, Aseltine, 1995) and to none of the other vari-ables; parents therefore apparently do not have very decisive role in defining students’ school experience. This result underscores how during this transi-tional phase, boys tend to branch out from their families, broadening their interactional horizons and focussing particularly on the support their class-mates and teachers provide (Santinello, Vieno, 2002). Conversely, both fami-lies and teachers are associated with school-related stress (Perry et al., 1993;


231

Samdal et al., 2000; Steptoe, 1991). Hence, in terms of the support, teachers can either improve school life for boys or, in negative terms (demands), make it worse. Conversely, boys seem to experience parental figures as a sort of complication, i.e., as someone over-burdening them with expectations.

Our results suggest that girls’ relationships with their schools involve as-pects that differ from those of their male counterparts. In addition to being a crucial factor in school satisfaction and achievement, teacher support of girls can be a protective factor against school-related stress (Voelkl, 1997). Parental support appears to be more decisive for girls than for boys in both school satisfaction and achievement. This result underscores how girls may be more sensitive than boys to this form of support, confirming a need in pubertal girls to feel supported by their parents and particularly, by their mothers (San-tinello, Vieno, 2002). The influence of parental and teacher demands takes on the same connotations in girls as for their male counterparts. Indeed, this variable is apparently negatively linked to all aspects of girls’ relationships with their schools.

Unlike for their male classmates, for whom classmate support is related to academic achievement, classmate support of girls seems to be associated with school satisfaction (Wentzel, Watkins, 2002). Thus, boys apparently rely on their classmates by informationally and instrumentally concretising their support (Brown et al., 1986). Conversely, girls seem to be more autonomous in their academic achievement, reserving another role for their classmates--one that is linked more to the pleasantness of school climate. We therefore suggest that girls see classmates as a form of emotional support, more so than practical (informational and instrumental) support.

The right-hand side of both models, which refer to the effect perceived school setting has on early adolescent well-being, showed several gender simi-larities. First, in agreement with various authors (Garralda, 1996; Hurrelman et al., 1995; Resnick et al., 1993; Samdal et al., 2000), the three well-being in-dicators are linked to school satisfaction for both girls and boys. This result highlights once more how truly important school settings are in early ado-lescents’ lives. In fact, when early adolescents repeatedly experience trouble at school, they also experience anxiety and even extreme distress, frequently with profound personal suffering (Sarason, 1980).

For boys, good school achievement affects their quality of life, whereas for


232

girls, it has a protective effect against substance use. This result underscores how boys’ and girls’ reaction strategies differ. If, on one hand, boys seem to react very directly and immediately to their not doing well in school, girls ap-parently turn to use substances as a means to cope with their particular situ-ation of stress and suffering. This result apparently contradicts experimental evidence (Barriga et al., 2002) attributing more behavioural reactivity to boys and a greater internalisation tendency to girls. We believe that our result may be accounted for by a cultural change occurring in northern Italy over the last two decades, calling for reaction strategies in both genders that were once exclusively part of the male domain.

One further gender difference concerns the repercussions of school-re-lated stress on student well-being. In fact, girls seem to be more sensitive to this school setting factor, by reacting with psychosomatic symptoms. Yet, it is also important to note the possibility that the association between these two variables may be inverted. Indeed, it is highly likely that students with more health problems experience greater difficulty in dealing with the developmen-tal tasks that are associated with their school settings, and they can thereby perceive greater school-related stress.

The main limitation of this study lies in the type of indicators that were used. First, only students’ perceptions were evaluated. Future research should rely on other informational sources, such as parents’ and teachers’ respons-es. Secondly, both the epidemiological nature of the study and international standardisation of the items ensured that several of the examined areas (qual-ity of life, achievement, and school-related stress) were measured by indica-tors (proxy) of the investigated constructs. Future research should re-test the model with more exhaustive scales. In conclusion, therefore, school life can deeply impact developmental processes (although in different ways), by pro-viding students with diverse occasions to try out their social and cognitive skills. Great commitment is required of Italian early adolescents to success-fully face the developmental tasks associated with the school-career choices they must make in transition from middle school to high school, and there-fore, this period can be particularly critical for them. This is precisely when their peers, but mostly their teachers, play a crucially supportive role--a role that cannot be confined to simply presenting facts and information about the world and asking students to memorise them.


233

REFERNCES

Allison, K.R., Adlaf, E. M., Mates, D. (1997), Life strain, coping, and sub-stance use among high school students. Addiction Research, 5, 251-272.

Argyle, M., Martin, M. (1991), The psychological causes of happiness. In F. Strack, Argyle, M. (Ed.) Subjective well-being: An interdisciplinary perspective. International series in experimental social psychology, Vol. 21 (pp. 77-100). Elms-ford, NY: Pergamon Press, Inc.

Aunola, K., Stattin, H., Nurmi, J. (2000), Adolescents’ achievement strategies. School adjustment, and externalising problem behaviour. Journal of Youth and Adolescence, 29, 289-306.

Barriga, A.Q., Doran, J.W., Newell, S.B., Morrison, E.M., Barbetti, V., Dean, B. (2002), Relationships between problem behaviors and academic achieve-ment in adolescents: the unique role of attention problems. Journal of Emotional and Behavioral Disorders, 10, 233-240.

Battistich, V.S.E., Watson, M., Solomon, D., Lewis, C. (2000), Effects of the Child Development Project on students’ drug use and other problem be-haviors. Journal of Primary Prevention, 21, 75-99.

Bonino, S., Cattelino, E. (2000), Il fumo di sigarette in adolescenza: implica-zioni per la promozione della salute. Psicologia della Salute, 1, 33-49.

Bost, K.K., Vaughn, B.E., Washington, W.N., Cielinski, K.L., Bradbard, M.R. (1998), Social competence, social support, and attachment: Demarcation of construct domains, measurement, and paths of influence for preschool children attending Head Start. Child Development, 69, 192-218.

Brown, B.B., Eicher, S.A., Petrie, S. (1986), The importance of peer group (“crowd”) affiliation in adolescence. Journal of Adolescence, 9, 73-96.

Bru, E., Stephens, P., Torsheim, T. (2002), Students’ perceptions of class management and reports of their own misbehaviour. Journal of School Psy-chology, 40, 287-307.

Butters, J.E. (2002), Family stressors and adolescents cannabis use: a pathway to problem use. Journal of Adolescence, 25, 645-654.

Byrne, B.M. (1989), Multigroup comparison and the assumption of equiva-lent construct validity across groups: methodological and substantive is-sues. Multivariate Behavioral Research, 24, 503-523.

Cattelino, E. (2000), Relazioni con i coetanei in adolescenza: Il contributo


234

degli amici e del partner nella promozione del benessere. Età Evolutiva, 65, 102-111.

Cohen, S., Syme, S.L. (1985), Social support and health. Orlando, FL: Academic Press.

Connell, J.P., Spencer, M.B., Aber, J.L. (1994), Educational risk and resilience in African-American youth: Context, self, action, and outcomes in school. Child Development, 65, 493-506.

Corbetta, P. (1992), Metodi di analisi multivariata per le scienze sociali. Bologna: Il Mulino.

Covington, M.V., Beery, R.G. (1976), Self-worth and school learning. New York: Holt, Rinehart and Winston.

Cudeck, R., Browne, M.W. (1992), Constructing a covariance matrix that yields a specified minimizer and a specified minimum discrepancy func-tion value. Psychometrika, 57, 357-369.

Dawson, D.A. (1998), Volume of ethanol consumption: Effects of different approaches to measurement. Journal of Studies on Alcohol, 59, 191-197.

DuBois, D.L., Burk-Braxton, C., Swenson, L.P., Tevendale, H.D., Lockerd, E.M., Moran, B.L. (2002), Getting by with a little help from self and oth-ers: Self-esteem and social support as resources during early adolescence. Developmental Psychology, 38, 822-839.

Fraser, B.J. (1994), Research on classroom and school climate. In D.L. Gabel (Ed.) Handbook of research on science teaching and learning (pp. 493-541). New York: MacMillan.

Galbiati, E., Santinello, M. (2002), La percezione dell’ambiente scolastico e il rapporto con la scuola nei preadolescenti. Psicologia Scolastica, 1, 297-309.

Garralda, M.E. (1996), Somatisation in children. Journal of Child Psychology and Psychiatry and Applied Disciplines, 37, 13-33.

Gore, S., Aseltine, R.H. (1995), Protective processes in adolescence: Match-ing stressors with social resources. American Journal of Community Psychology, 23, 301-327.

Graham, P., Stevenson, J., Flynn, D. (1997), A new measure of health-related quality of life for children: Preliminary findings. Psychology and Health, 12, 655-665.

Harter, S. (1996), Teacher and classmate influence on scholastic motivation, self-esteem, and level of voice in adolescents. In J. Juvonen, K.R. Ventzel


235

(Ed.), Social motivation: Understanding children’s school adjustment (11-42). New York: Cambridge University Press.

Hosman, C.M.H. (2000), Prevention and health promotion on the interna-tional scene: The need for a more effective and comprehensive approach. Addictive Behaviours, 25, 943-954.

Hoagwood, K., Jensen, P.S., Petti, T., Burns, B.J. (1996), Outcomes of mental health care for children and adolescents: I. A comprehensive conceptual model. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1055-1063.

Huebner, E.S., Laughlin, J.E., Ash, C., Gilman, R. (1998), Further validation of the Multidimensional Students’ Life Satisfaction Scale. Journal of Psych-oeducational Assessment, 16, 118-134.

Hurrelmann, K., Leppin, A., Nordlohne, E. (1995), Promoting health in schools: The German example. Health Promotion International, 10, 121-131.

Idler, E.L., Benyamini, Y. (1997). Self-rated health and mortality: A review of twenty seven community study. Journal of Health and Social Behaviour, 38, 21-37.

Institute of Medicine (1994), Reducing risk for mental disorder: Frontiers for preven-tive intervention research. Washington, DC: National Academy Press.

Jessor, R. (1991), Risk behaviour in adolescence: A psychological framework for understanding and action. Journal of Adolescent Health, 12, 597-605.

Jöreskog K. G. (2000), Interpretation of R2 Revisited. Disponibile [on line]: http://www.ssicentral.com/lisrel/column5.htm.

Jöreskog, K.G., Van Trillo, M. (1973), Lisrel: a general computer program for esti-mating a linear structural equation system involving multiple indicators of unmeasured variables. Research Report 73-5. Uppsala University.

Kazdin, A.E. (1993), Psychotherapy for children and adolescents: Current progress and future research directions. American Psychologist, 48, 644-657.

Knishkowy, B., Palti, H., Tima, C., Adler, B., Gofin, R. (1995), Symptom clus-ters among young adolescents. Adolescence, 30, 351-362.

Mortimore, P. (1995), The positive effects of schooling. In M. Rutter (Ed.), Psychological disturbance in young people: Challenge for prevention (pp.333-366). Cambridge: Press Syndicate of the University of Cambridge.

Nutbeam, D., Smith, C. Moore, L., Beauman, A. (1993). Warning! Schools can damage your health: Alienation from school and its impact on health


236

behaviour. Journal of Paediatric and Child Health, 29, 25-30.Perry, C.L., Kelder, S.H., Komro, K. A. (1993), The social world of adoles-

cents: Family, peers, schools, and the community. In S.G. Millstein, AC. Petersen, E.O. Nightingale (Ed.), Promoting the health of adolescents: New directions for the twenty-first century (pp. 73-96). London: Oxford University Press.

Resnick, M.D., Harris, L.J., Blum, R.W. (1993), The impact of caring connect-edness on adolescent health and well-being. Journal of Paediatric and Child Health, 29, 3-9.

Roeser, R.W., Eccles, J.S., Sameroff, A.J. (1998), Academic and emotional functioning in early adolescence: Longitudinal relations, patterns, and pre-diction by experience in middle school. Development and Psychopathology, 10, 321-352.

Roeser, R.W., van der Wolf, K., Strobel, K.R. (2001), On the relation be-tween social-emotional and school functioning during early adolescence: Preliminary findings from Dutch and American samples. Journal of School Psychology, 39, 111-139.

Samdal, O., Nutbeam, D, Wold, B., Kannas, L. (1998), Achieving health and educational goals through schools--a study of the importance of the school climate and the students’ satisfaction with school. Health Education Research, 13, 383-397.

Samdal, O., Wold, B., Torsheim, T. (2000). The relationship between stu-dents’ perception of school and their reported health and quality of life. In WHO (Ed.), Health and health behaviour among young people. WHO Policies series, 1. [On Line] Disponibile: www.who.dk/hbsc.

Santinello, M., Vieno, A. (2002), L’influenza del sostegno sociale ed econo-mico sul disagio dei preadolescenti. Bollettino di Psicologia Applicata, 238, 33-43.

Sarason, I.R. (1980), Test anxiety: Theory, research and application. Hillsdale: Earl-baum.

Steptoe, A. (1991), The links between stress and illness. Journal of Psychosomatic Research, 35, 633-644.

Tani, F. (2000), Le amicizie negli anni della scuola: Un’indagine sulle caratteri-stiche personali dei bambini con molti amici. Età Evolutiva, 67, 52-59.

Thompson, S.K. (2002). Sampling. Second Edition. New York: John Wiley &


237

Sons.Torsheim, T., Wold, B. (2001), School-related stress, school support, and so-

matic complaints: A general population study. Journal of Adolescent Research, 16, 293-303.

Torsheim, T., Wold, B., Samdal, O. (2000), The Teacher and Classmate Sup-port scale: Factor structure, test-retest reliability and validity in samples of 13- and 15-year-old adolescents. School Psychology International, 21, 195-212.

Voelkl, K.E. (1997), Identification with school. American Journal of Education, 105, 294-318.

Wasserstein, S.B., La Greca, A.M. (1996). Can peer support buffer against behavioral consequences of parental discord? Journal of Clinical Child Psy-chology, 25, 177-182.

Wentzel, K.R. (1998), Social relationships and motivation in middle school: The role of parents, teachers, and peers. Journal of Educational Psychology, 90, 202-209.

Wentzel, K.R., Watkins, D.E. (2002). Peer relationships and collaborative learn-ing as contexts for academic enablers. School Psychology Review, 31, 366-377.

WHO (1998). Health Behaviour in School-aged Children: research protocol from the 1997-98 survey. Edinburg: University of Edinburgh.

Wold, B., Aaroe, L.E., Smith, C. (1994). Health behaviour in school-aged children. A WHO cross-national survey (HBSC). Research protocol for the 1993-94 study. Hemil-Report, 14.

ABSTRACT

This study investigates the relationship between school climate and well-being in early adolescence testing a causal theoretical model (Samdal et al., 2000). A representative sample of 1,347 students (48.6% female) from 82 middle schools filled out a questionnaire based on the WHO-coordinated HBSC1 survey. Results revealed how perceived school climate and the de-

1 The HBSC study in Veneto is funded by Regione Veneto (D.G.R. n° 203 of April, 4, 2000). HBSC is a cross-na-tional study co-ordinated from the European Office of WHO. International co-ordinator for the 1997/99 study: Candace Currie, University of Edinburg; Data Bank Manager: Bente Wold, University of Bergen; National co-ordi-nator: Franco Cavallo, University of Torino.


238

gree to which students experience it as positive or negative can influence not only their motivation and school satisfaction, but also their lifestyles, health, and quality of life. Overall, the results of this study confirmed the proposed theoretical model, revealing gender differences, and confirming the impor-tance of the role of teachers in conditioning both school climate and the bio-psychosocial well-being of early adolescents. In conclusion, this study provides experimental evidence that contributes to a better understanding of the social processes determining early adolescent well-being.

Received March 2 2004; accepted September 10 2004

Acknowledgements We are grateful to Isaac Prilleltensky (Vanderbilt University) for his comments on the

earlier drafts.


239

CHILDREN’S DEFINITIONAL SKILLS AND THEIR RELATIONS WITH

METALINGUISTIC AWARENESS AND SCHOOL ACHIEVEMENT

Gianluca Gini*, Beatrice Benelli*, Carmen Belacchi**

INTRODUCTION

Linguistic definition is a kind of judgement that either establishes a new piece of knowledge as in the case of a new scientific discovery (institutive definitions; e.g., “radio waves”) or makes already existing and shared infor-mation explicit (lexicographic definitions; e.g.: the definition of “apple” or “beauty”). Watson (1985) defined the nature of definitions considering four central points. First of all, definitions have a conventional linguistic form, which states semantic equivalence between the definiendum (the word to be defined) and the definiens (the linguistic structure used to explain it). More precisely, the internal structure of the definiens has a conventional form, that is a superordinate category term (introduced by the copula) and followed by differentiating characteristics (Anglin, 1977) (“An X is an Y that Z”).

Moreover, definitions are conventional, rather than “true”, word mean-ings, in that they are not sufficient to circumscribe meaning across all speak-ers, whereas they reflect “how a word has been used historically” (Watson, 1985, p. 184).

As a third point, definitions represent explicit expressions of word mean-ings that remain largely implicit in oral language: the process of definition serves to make explicit the meaning of a word which has been acquired in

* Development and Socialization Psychology Department, University of Padova – Italy.** University of Urbino – Italy.

Gianluca Gini, Beatrice Benelli, Carmen Belacchi

240

discourse and it is usually used for “pragmatic”, rather than linguistic pur-poses.

Finally, Watson (1985) underlined the metalinguistic nature of definitions, which require a reflection upon the properties and uses of language, because they establish relations between parts of language (the definiendum and the definiens, the word and its superordinate category) rather than parts of world.

The development of definitional skills. Many developmental studies (Benelli, Arcuri & Marchesini, 1988; Davidson, Kline & Snow, 1986; Johnson & Ang-lin, 1995; Snow, 1990) have shown that in answering questions such as: “What is a (stimulus word)? What does it mean?” younger children – up to 5/6 years – tend to give mostly descriptions of objects (using the HAS structure: “a dog has four legs”) whereas older children – starting from the approximate age of 7 – tend to add superordinate categorical terms, showing the so-called ISA structure: “a sofa is a long and comfortable piece of furniture”. The shift from the most primitive formulae to the canonical ones takes the form of different developmental levels, each characterized by a typical cognitive-lin-guistic approach on the part of the child (Litowitz, 1977; Watson, 1985).

This developmental process leads children at the end of elementary school to achieve the structure of metalinguistic definitions, in which they must combine correct and informative contents with appropriate, i.e. canonical, forms. Learning to combine form and content is not a simple process, which probably explains why it takes time for children to accede to the most so-phisticated definitional forms, as several studies have shown both with cross sectional (e.g. Watson, 1985) and longitudinal techniques (Kurland & Snow, 1997).

The metalinguistic nature of definitions. Defining words can be considered a metalinguistic task (Snow, 1990; Wehren, De Lisi, & Arnold, 1981), even though not all types of definition reflect high levels of metalinguistic ability, for example ostensional and extensional responses (pointing at, or listing, instances of the definiendum and of its sensory properties) simply require a referential use of language to describe characteristics of obj.ects or events (McGhee-Bidlack, 1991) whereas metalinguistic definitions require an use of language which is auto-referential. This involves using language to explain language itself, to make its formal and content components explicit, to reach the semantic core of a linguistic item, by means of other linguistic expres-sions. Moreover, the shift from the simplest forms of definition to better

Children’s definitional skills and their relations with metalinguistic awareness

241

developed answers not only requires a good knowledge of the concept to be defined but also, and above all, the awareness of what a definition is and what it is used for (Watson, 1985).

An important metalinguistic component in the process of defining a word is the ability both to systematically analyse the relevant information about a concept, in order to select the most discriminative contents, and to self-moni-tor the linguistic forms used to express them (Snow, 1990; Snow, Cancino, De Temple, & Schley, 1991); more precisely a strong relation exists between definitional skills (particularly the tendency of using superordinate terms) and the conception of language as a conventional instrument of transmis-sion of information about the world (Benelli, 1988). Only if the child realises that names are arbitrary means used to refer to objects, i.e. the word-referent differentiation (Tulviste, 1993), he/she can consent to use different linguistic labels (synonyms or superordinates) to denote the same object.

Definitional skills and school experience. Several authors suggested that defi-nitional skills, as a measure of decontextualized language abilities, are close-ly related to school success, in general, and the development of literacy, in particular (Chall, 1987; Snow, 1990; Snow, Cancino, Gonzalez, & Shriberg, 1989). Definitional skills and school experience have more than one point in common, in a reciprocal and bi-directional relation. In fact, children gradu-ally learn both the correct formal structure of the adult-like definitions and its content constraint (that is selection of the most relevant information) through examples of “good definitions” they can experience at school, i.e. in teachers’ speeches and in text-books, even when definitions are not the real issue of teaching (Snow, 1990).

In this process, the role of adults is fundamental, in that they have the possibility to explicitly provide the conventional structures of language cor-recting, expanding and elaborating spontaneous productions of children (Watson, 1985). This general mechanism works also in the case of linguistic definitions when, for example, teachers encourage pupils to analyze the more abstract aspects of concepts, introducing superordinate terms and the use of more complex definitional formulae (Benelli, Arcuri, & Marchesini, 1988; Snow et al., 1989).

Finally, schooling and some of its specific aspects, such as the acquisition of reading and writing, induce a more detached and reflective attitude towards language in general and more formal and complete ways of word defining


242

(Sinclair, 1986). Children who are learning literacy, in fact, are often required to manipulate word meaning and to use their lexical knowledge to provide definitions for words they read (Gutierrez-Clellen & DeCurtis, 1999).

Therefore, as regards the first direction of the relationship between school experience and definitional skills, different authors agree that school expo-sure and the opportunity to practice formal definitions in classroom do influ-ence children’s word definitional skills.

The ability to define may relate to the development of a “literate register” mediated by the child’s literacy experience (Kikas, 1993). The focus on for-mal definitions is more typical of school environments than of other settings and during school activities formal definitions are emphasized to develop new vocabulary. For example, Snow (1990) found that school exposure was related to children’s performance on definition tasks more than their home experience. Watson and Olson (1987) concluded that schooling experience is crucial in the development of the ability to define words.

However, the inverse relationship is also true, in that definitional skills can positively influence school achievement in children of different levels of schooling. This relation is based on the fact that academic tasks, such as reading and writing, at least starting from upper elementary school, cannot be achieved without a good mastery of decontextualized language skills, both in its productive and receptive forms. Empirical evidences of this linkage were found by Snow and colleagues (1989) who analyzed the performance of elementary school children in a definitional task. Their results showed that only children’s definitional score in formal definitions (i.e. those answers which contain the copula and a superordinate term) was positively related with the reading and language scores in the California Achievement Test, whereas their informal definitional score was not.

The attainment of a good level of definitional skills and the development of different components of this ability represent a valid means of compre-hension and learning for pupils. In other words, definitions can be used to stipulate meaning, when “the semantic structure of a lexical item is entire-ly contained within the linguistic expression that constitutes the definiens” (Watson, 1985, p. 195); this role is particularly important within particular arguments or contexts, such as scientific subjects, where discourses refer to highly abstract concepts, for example mathematical or geometrical concepts (“A circle is the locus of points equidistant from a given point”). Through


243

classroom interactions and practice, pupils learn to describe their scientific knowledge using formal definitions.

Moreover, definitional skills predict the acquisition of literacy better than the general knowledge of word meaning (Olson, 1991; Snow et al., 1989). Snow et al. (1989), in particular, stated that the relations between children’s vocabulary size and reading skill are not based only on children’s knowledge about word meaning, but also on how they use this knowledge, i.e. their meta-linguistic awareness about language and its uses. Considering that children’s production of a good definition is a reliable evidence of his/her knowledge of a particular word (Johnson & Anglin, 1995), many researchers believe that definitions can be used as a measure of children’s vocabulary acquisition, which in turn is a strong predictor of reading ability (Anderson & Freebody, 1983; Nagy & Herman, 1987).

The different cognitive, linguistic, metalinguistic and communicative abili-ties involved in the process of word definition are the same competencies which everyday learning tasks demand to all schoolchildren. Therefore, con-cluding this analysis, it is possible to hypothesize that definitional skills can be an ability positively related to school success (Guiterrez-Clellen & De Curtis, 1999). More in general, several studies on definitions found positive relations between children’s definitional skills and intelligence, measured through stan-dardized tests such as, for example, the Stanford Binet (M – L) Scale (Storck & Looft, 1973) or the Test of Word Knowledge (TOWK) or the Test of Language Development-2 (Told-2) (see Nippold, 1995 for a review). How-ever, these instruments were used to establish general norms for definitional development, more than to understand the processes underlying the acquisi-tion of definitional skills.

Other authors (Benelli, Belacchi, & Lucangeli, 2004) confirmed this positive relation by means of a standardized intelligence test, the Wechsler’s scales. In this study, in fact, all the intelligence measures did correlate with the ability to give definitions: the higher were all the intelligence indexes (Verbal I.Q., Performance I.Q. and Total I.Q.), the better was definitional compe-tence. Such data support the idea that definitional skills can be accounted for both in terms of general intellectual abilities (as revealed by the three I.Q. measures) and in terms of more specific abilities, such as lexical knowledge (Vocabulary scale) or ability to identify relations of numerical-mathematical (Arithmetic scale), spatial (Figure completion scale and Cube drawing scale)


244

and logical-categorical nature (Similarities). Also metalinguistic ability has been shown to correlate with cognitive de-

velopment in general, and metacognition in particular, literacy skills and oral language skills (Hakes, 1980; Menyuk, 1984; Smith & Tager-Flusberg, 1982). Moreover, metalinguistic awareness develops as a function of schooling (Fer-reira & Morrison, 1994) and its relationships with school performance in different kind of tasks have been confirmed many times (Borkowski, Ryan, Kurtz, & Reid, 1983; Olson & Astington, 1993; Schneider & Naeslund, 1993). In particular, different studies evidenced the positive effects of a good level of metalinguistic awareness on pupils’ different academic achievements, such as learning of reading and writing, their ability in decontextualized language tasks, both in monolingual and bilingual children (Carlisle, Beeman, Davis, & Spharim, 1999; Dreher & Zenge, 1990; Francis, 1999).

Aims of the study. The present study started from the consideration that even though some authors have explicitly stressed that definitional skills include metalinguistic components (McGhee-Bidlack, 1991; Snow, 1990; Watson, 1985; Wheren et al., 1981) they put forward this claim as a theoreti-cal, conceptual starting point more than as empirical evidence of relations between definitional competence and independent measures of metalinguis-tic skills. Moreover, there are no studies in literature that considered together the relations between definitional skills, metalinguistic awareness and school success.

Therefore, the general aim of this study is to empirically analyze (a) the relations between children’s definitional skills and their metalinguistic aware-ness measured through an independent task; (b) the relations between defi-nitional ability and school achievement in Italian elementary school children; (c) the relations between metalinguistic awareness and school success.

School achievement is assessed through two different measures. The first measure are children’s school marks in subject-matters such as Italian lan-guage and mathematics. These two specific matters have been selected be-cause they represent two different cognitive abilities, the former more liter-ary-linguistic and the latter more logical-mathematical. We did not consider other subject-matters, such as science or history, because they are more ori-ented on disciplinary contents. The second measure of school achievement are four aspects of children’s linguistic-expressive competence, considered crucial for both school success and definitional skills. Specifically, the four


245

competencies we considered are: Lexical knowledge, Syntactic competence, Communicative competence and Role taking ability. All these competencies can be considered strongly related with school achievement, because they are all abilities required to successfully deal with several school tasks and requirements. Therefore, teachers were asked to rate their children on these competencies, starting from their familiarity with them and their teaching experience.

As regards the relations between these competencies and definitional skills, the first two abilities (lexical knowledge and syntactic competence) are directly related with the formal aspect of definitions (the knowledge of the definiendum and the mastering of the morpho-syntactic rules of the definiens), whereas the other two competencies are related to the content and communi-cative aspects of definitions, that is the selection of the most relevant content for the identification of the referent.

In order to allow the differentiation between linguistic-formal and con-tent-communicative aspects of definitions, children’s answers will be scored according to two different Scales, one which stresses formal aspects, that is the morpho-syntactic rules, of definitions (the so-called Formal Definitional Scale) and the other more oriented on content, that is the informative power of definitions, independently from formal accuracy (the Content Definitional Scale). These scales can be considered two different, even though strongly related, ways to evaluate children’s definitions. As we already said, all defini-tions are formed of both formal and content components (Watson, 1985). Whereas the first one refers to the conventionally established form of the definiens, in other words the Aristotelian format, the second component refers to the communicative efficacy of definitions, that is the extent to which they allow identification of the “intended referent” (Watson,1995). We believe that relying only on the complete mastery of formal rules, in order to evaluate the quality of definitions, might underestimate the communicative strength of some kinds of children’s definitional answers. More precisely, natural, or pragmatic, definitions (Rey-DeBove, 1971) are very representative of young children definitional ability: sentences such as “hens lay eggs and cackle” may be as useful in pinpointing their referent as their corresponding metalinguis-tic, or prescriptive, definition: “hens are animals that lay eggs and cackle”. Clearly the latter definition is much better in every respect, but even the for-mer can be useful and appropriate, according to the communicative context


246

(Benelli et al., 1988). In other words, even younger children may acknowl-edge the importance of being perspicuous (Grice, 1975) in their messages; simply, they follow this communicative rule according to their most typical linguistic-cognitive approach: up to the approximate age of six, the dominant representational system has still a narrative nature (Bruner, 1986; Nelson, 1996). Considering the possible relations among these variables we can ex-pect that a) as suggested in literature, definitional skills and metalinguistic awareness are positively related even at the beginning of elementary school (Snow, 1990; Wehren et al., 1981); b) definitional skills and school success are positively related because mastering formal definitions can be considered a result of schooling and, at the same time, a requisite for school achievement (for ex., Snow et al., 1989); c) metalinguistic awareness and school success are positively related because metalinguistic competence develops as a function of schooling and, in particular, of literacy acquisition (for ex., Ferreira and Morrison, 1994).

METHOD

ParticipantsForty-six 6-years-old children (mean age = 6 years and 8 months) and

forty-two 10-years-old children (mean age = 10 years and 7 months) attend-ing an Italian public elementary school participated in the study. Both groups were equally divided between boys and girls. These two age-groups were cho-sen as representative of two very different levels of schooling, the beginning and the end of Italian elementary school.

The SES background of children’s families varied from low to medium. In terms of racial/ethnic background, all children in the sample were White-Eu-ropean children. All children received school and parental permission. Chil-dren with language or cognitive impairment were excluded from the sample.

Materials and procedureDefinition task: 24 words, belonging to three grammatical categories, were

used as stimuli: 8 nouns (ability, clown, donkey, kindness, orange, rivalry, spy-


247

ing, umbrella), 8 adjectives (blonde, contagious, innocent, polite, risky, round, smooth, thin) and 8 verbs (to baffle, to burn, to connect, to emigrate, to force, to hit, to think, to tolerate). Within each category, half items were con-crete, or perceptually based (e.g. clown, blonde, to hit), whereas the others were abstract, or less sensorially based (e.g. kindness, innocent, to think). The stimulus list was formed by selecting the items from the Italian version of the Stanford-Binet Scale and the Wechsler Scale for Children. Children were individually interviewed in a quiet room and asked to answer the following question: “What does the word X mean?”. Their definitions were verbatim transcribed and the same responses were judged on two different five-level scales: the so-called “Formal Definitional Scale” and the “Content Defini-tional Scale”. These two scales should emphasize different aspects of the same general definitional ability.

The Formal Definitional ScaleThis scale puts particular emphasis on the role of formal, i.e. morpho-

syntactic rules in the development of children’s definitional skills: the more correct and syntactically organised the sentence, the higher is the definitional level. This does not mean that content is irrelevant; it simply means that well shaped definitional formulae could reveal more about children’s awareness of the canonical format, even if its content is incorrect, than an appropriate content expressed in simple, descriptive ways (ex. muzzle “a donkey has a muzzle”). For example, a definition such as “a board game requiring to form pictures from scattered small pieces” used to explain the word “muzzle” sug-gests that the child possesses the format of the formal definition even if he/she probably misunderstood the word.

Actually, a stimulus word can be misunderstood, thus giving rise to a wrong answer, but sophisticated definitional sentences, anyhow, require at least a minimal analysis of what a definition structure is, or should be.

The five-level structure of the Scale can be described as follow:• level I (Non-definitional): in this level no verbal answers are provided

(Score 1);• level II (Pre-definitional): single-word answers. Linguistic answers are

produced, but in very simple forms: a single word follows the stimulus item, linked to it by:


248

A) phonetic similarity, or rhymes (muzzle puzzle); B) semantic associations (muzzle dog). (Score 2);

• level III (Quasi-definitional). This third level shows an initial use of mor-pho-syntactic devices, in order to articulate and expand the semantic content. Such devices are: A) simple prepositions added to the stimulus word (such as muzzle of a dog: clown in a circus); B) Linguistic structures called "periphrastic attempts", which consist of longer ut-terances introduced by temporal, spatial, relative, etc. connectives (such as: muzzle – where the dog has its teeth; orange – that can be eaten or squeezed). The resulting structures, in any case, cannot be considered as real clauses, i.e. complete and self-sustaining utterances, but only "pieces of sentences" (Rey-DeBove,1971) (Score 3);

• level IV (Definitional non integrated). The formal property characteriz-ing this fourth level is the appropriate linguistic structure, which begins to be shaped as autonomous and correct clauses with the conjugated verb. At this level content can be still not adequate.

The definitional structure at this fourth level takes basically three forms:

A) Perceptual-Narrative Periphrases: descriptions of objects’ senso-rial properties or narratives of events, such as: orange it is round and juicy; or you can squeeze it; or even: oranges grow on trees.

The resulting sentences are complete, even though verbs are mostly action verbs or the copula used to describe things (“the apple is red”);

B) Simple Categorization: in this second form, where the initial Ar-istotelian structure ("an X is a Y") is used, the copula begins to play the semantic correspondence role: the term Y is either a superordinate or a synonym (orange it is a piece of fruit; "mean" means nasty). However no specifications are added, the function of which is to help identify the referent;

C) Categorizing Periphrases Inadequate in Content: the linguistic structure is the most evolved (i.e. specified) Aristotelian formu-la, but the content is somehow inappropriate: it can be either correct but not sufficiently informative to allow for referents' identification (orange it is a round piece of fruit); or correct


249

but partial and limited, such as in idioms and metaphoric usage (clown means that you are stupid); or even totally wrong ("The ambergris is a mineral fossil") (Score 4);

• level V (Definitional Integrated). At this last level, both form and con-tent are correct and integrated with each other. The structure is mostly formed of: copula + introductory term (category or abstract) + speci-fication, with no repetition of the stimulus word. The introductory terms of the definiens can be as follows:

A) general place-holders such as: “thing”, “something”, “an object”, “someone”, etc. (“a clown is somebody who makes you laugh in a circus”;);

B) natural category terms, such as “instruments”, “animals”, “ve-hicles”, etc. (“A clown is a funny character you can see in circuses or shows”);

C) abstract relational terms such as: “place” “part of ”; etc. “condi-tion” etc. (a flower bed is a place where you can put and grow plants”; “the muzzle is the front part of an animal’s head”; “haste is a condition in which you have not enough time to do things”);

D) items such as: “a noun”, “an adjective”, “the synonym of ”, etc., that is, items which make the linguistic/grammatical category of the stimulus word explicit (“innocent is the opposite of guilty”) (Score 5).

The Content Definitional ScaleThis second scale has the same five-levels structure of the previous scale

and was developed to measure the effectiveness of a definition from a con-tent-communicative point of view, using classification criteria which are rela-tively independent of formal correctness.

So, the Content Scale focuses on the effectiveness of the information pro-vided by children about a word meaning, allowing the interlocutor to identify the correct referent. In this scale lower scores were given to those answers which have the lowest informative power, whereas higher scores were given to those transmitting the clearest information, regardless of formal appropri-ateness, at least to some extent.


250

• Level I: Non Verbal Answers. Absence of answer (or “I don’t know”) or ostensional answers (Score 1);

• level II: Non Informative. This level include all the tautological defini-tions, that is answers giving no new information but simply repeat-ing the definiendum with more or less linguistically complete structure (“blondish” for “blonde” as well as “a clown is someone who acts as a clown”). Since the non repetition of linguistic items acting as definien-dum in the definiens (paraphrase) is a basic rule of definitions, if this does not happen, no real information is actually added. Also associative answers which simply hint at the semantic content of the to-be-defined concept (clown – circus; hen – egg) are put in this same level. Finally, level II includes answers with a completely wrong content, whatever the form may be. Clearly in this last case no adequate conceptual rep-resentation of the definiendum can be activated by the definiens. All these kind of answer don’t allow to reach the meaning of the word (Score 2);

• level III: Descriptive non discriminating. These answers provide semantically correct but insufficient information which, therefore, doesn’t allow to discriminate the “intended referent” from similar categorical exemplar or related concepts; their linguistic form may be either syntactically incomplete (clown – “when you go to the circus”) or correct and nar-rative (hen – “it lives in farms”). These kinds of answer allow only a partial access to the meaning of the word (Score 3);

• level IV: Categorical non discriminating. At this level, definitions have a good formal structure containing mostly superordinate terms; this is why they reach a higher definitional level, being superordinates the most rel-evant information (Watson, 1995) that is, allowing inferences about the possible nature of the stimulus. However these answers are not totally adequate from a communicative point of view, because no specifica-tions are given or, if so, they are again not discriminating (rivalry – “a frequent condition among people”) (Score 4);

• level V: Informative and discriminating. The last level of the scale includes definitions which reach the best communicative efficacy through, of course, a syntactically correct and complete sentence, which is a neces-sary prerequisite to convey an informative content. These definitions can be formed of:


251

A) really effective perceptual-narrative descriptions (“a clown is fun-ny and makes you laugh at the circus”);

B) formally correct and informative categorizations (clown is a person disguised who makes you laugh in circuses or shows) (Score 5).

As regards the first scale, judges were asked to evaluate if a definition was correct according to the morpho-syntactic rules of the conventional defini-tional formulae (the Aristotelian format), whereas for the second scale they were asked to evaluate if the content of children’s answers was sufficient to identify the correct referent, that is the definiendum.

Interjudge reliability, calculated from a random sample of 22 children (about 20% of the total sample), was 91% for the Formal Scale and 94% for the Content Scale. Disagreements were resolved through discussion.

Metalinguistic Test: the metalinguistic test aims at measuring children’s knowledge about language and its use, by identifying different components responsible for a mature metalinguistic approach. The test is made up of four parts, which analyze different aspects of the metalinguistic competence: 1) Lexical awareness (questions 1-2): these items require a reflection about the nature of words (ex., “Is “ball” a word or not?”). For their formulation we referred to Benelli (1988); 2) Concept of definition (questions 3-4): these items evaluate a particular kind of metalinguistic knowledge which refers to lexicographic definition of words, that is the possibility to use language to explain language (ex., “Do you know what a definition is? What does it mean to define a word?”); 3) Relations between words (question 5): this section measures three different kind of relations between words: synonymy (ex., “Could a cat be called a kitten?”), name exchangeability (ex, “Could a cat be called an horse?”) and superordination (“Could a cat be called an animal?”). For the formulation of these items we referred to Benelli (1988); and, 4) Ori-gin of names (question 6): it recalls the Piagetian questions about the origin of names for things (“Where do names of objects come from?”).

The internal reliability was calculated by means of the Cronbach’s Al-pha, which reached 0.74. For each question children received a score on a four-point scale (1-4) which considers not only the correctness of answers but also the type of explanation provided by children. The most developed explanations were those referring to language as an object of thought rather than simply a way to communicate (for ex., conceiving the nature of words


252

as having meanings even though arbitrarily established). The total score obtained by participants in the Metalinguistic Test was

calculated by summing the score in each of the 21 questions of the question-naire (range: 21-84). The full version of the test and the coding criteria are reported in Appendix.

School Marks: school-marks, attributed to children by their teachers in Ital-ian language and mathematics respectively, were collected. These two differ-ent subject-matters represent different but related abilities: the former based on linguistic appropriateness, richness and fluency, the latter based more on abstract, clearly relational but still linguistically coded abilities. Marks in Ital-ian elementary school-system are usually distributed on 5 points (from insuf-ficient to very good).

Teachers’ Ratings: teachers were asked to rate the quality of linguistic and communicative competence of each pupil along four dimensions: 1) Lexical knowledge; 2) Syntactic competence; 3) Communicative competence; 4) Role taking. The instructions for teachers were as follow: “Please, evaluate the lexical knowledge (syntactic competence, etc.) of each pupil in your class on a four point scale, from insufficient to very good”.

RESULTS

Preliminary analyses on the measures used in the studyFirst of all, we considered the relations between the two Definitional Scales

used to assess children’s definitional skills. A Correlational Analysis between the mean scores in the two Definitional Scales was performed: as we could expect, this analysis confirmed a high level of correlation between them (r = .826, p<.001), in that they measure two different components of the same general definitional ability. As regards the two measures of school success, another Partial Correlation Analysis, controlled for age, was performed in order to see if the four dimensions rated by teachers can be considered abili-ties really involved in school learning. Correlation coefficients, reported in Table 1, show a strong relation between school marks and teachers’ ratings.


253

The first measure represents a more “objective” and conventional evaluation of school success, focused on disciplinary acquisitions, whereas the second one refers to more general dimensions, involved in many aspects of learning and development.

Lexical Knowledge

Syntactic Competence

Communicative Competence

Role Taking

Marks in language .794* .795* .625* .697*

Marks in math .650* .743* .448* .587*

* p<.001 (two-tailed test).

TABLE 1. Patial Correlations between Teacher’s ratings and School marks

Analysis of the definitional skills through the Formal ScaleFirst of all, it emerges that no-responses (Level I) are more frequent in 6-

year-old children (35%) than in 10-year-olds (10%). Percentages of different kinds of answers, presented in Table 2, clearly show an age-related difference: 10-year-old children have their modal response in Level V (42%), followed by Level IV (30%), whereas 6-year-old children have the major proportion of verbal answers in Level IV (34%), and Level III (20%) as the second major proportion.

Level I Level II Level III Level IV Level V Total

6-year-old (n=46)

39035%

938%

21620%

37834%

273%

1104100%

10-year-old (n=42)

10210%

162%

16216%

30630%

42242%

1008100%

Qa 19.74** 9.15** 3.30* 2.72 -33.65**

aQ: Tukey-test for independent proportions;significance levels: ** <.001; * <.05;critical Q values: for alpha <.05: Q = 2.77; for alpha <.001: Q = 4.05.

TABLE 2. Distribution of answers in the two age-groups: Formal Scale


254

For each level of the Formal scale, proportions of answer given by 6-year-old children and 10-year-olds were compared by means of a non-parametric test, the Tukey-test for independent proportions (Williams, 1992). Results clearly show that younger children produced a proportion of “non-defini-tional” answers significantly higher than older children, who performed bet-ter in the fifth level of the scale, i.e. definitions which reach the Aristotelian form (superordinate/abstract term plus specifications).

Contrary to what we may expect, the comparison between the two groups is not significant at Level IV. This result can probably be explained by the complex internal articulation of this level, which includes different kinds of answers, that is both descriptive-narrative definitions (Sub-category IVA) and categorizing definitions (Sub-categories IVB and IVC). In fact, if we consider these different kinds of definitions within the fourth level, it emerges that the major proportion of answers given by 6-year-old children are Perceptual-Narrative Periphrases (IVA = 58% within Level IV), whereas in 10-year-old children definitions are distributed in a similar way across the three sub-cat-egories of Level IV, with a small prevalence of Categorizing Periphrases In-adequate in Content (IVC answers = 40% within Level IV), thus revealing the emergence of the new categorizing approach.

Analysis of the definitional skills through the Content ScaleAs we have seen in the Method section, the Content Scale considers defi-

nitions stressing their content quality, i.e. their effectiveness from a commu-nicative point of view. The distribution of answers scored according to the criteria of the Content Scale is shown in Table 3.

Level I Level II Level III Level IV Level V Total

6-year-old (n=46)

39036%

39236%

19417%

1019%

272%

1104100%

10-year-old (n=42)

10210%

11612%

20120%

14314%

44644%

1008100%

Qa 20.40* 18.37* -2.45 -5.00* -36.96*

aQ: Tukey-test for independent proportions;significance levels: * <.001.Critical Q values: for alpha <.001: Q = 4.05

Table 3. Distribution of answers in the two age-groups: Content Scale


255

Six-year-old children show clear difficulties in producing truly informative definitions, as appears from the high proportion of Level II answers (36%), which include mostly tautological and associative answers and few completely wrong answers. Instead, the modal level of older children is Level V (44%), that is formally correct and fully informative definitions, with or without cat-egorization. The Tukey-test performed on the five levels of the Content Scale confirmed the significant difference in the level of informativeness between definitions given by 6-year-old and 10-year-old children.

Distribution of answers in the Metalinguistic TestThe distribution of answers produced by children in the Metalinguistic

Test is presented is Table 4. The comparison of the proportion of answers in each level performed through the Tukey-test clearly show that older children produced significantly more answers than younger children at the best meta-linguistic level (Level IV), that is those answers that reveal a better reflection about language as an object of thought, rather than simply as a tool for com-munication.

Level I Level II Level III Level IV Total

6-year-old (n=46) 9310%

27829%

40542%

19019%

966100%

10-year-old (n=42) 11113%

14216%

22425%

40546%

882100%

Qa - 2.79* 9.32** 10.77** - 17.46**

aQ: Tukey-test for independent proportions;significance levels: * < .05; ** <.001;critical Q values: for alpha <.05: Q = 2.77; for alpha <.001: Q = 4.05.

Table 4. Distribution of answers in the Metalinguistic Test

Relations between definitional skills and metalinguistic awarenessThe relations between definitional skills and metalinguistic ability were

analyzed in each age-group separately through a Correlational Analysis per-formed on children’s total score in each of the two Definitional Scales and


256

their score in the Metalinguistic Test. In Table 5, Pearson’s correlation coef-ficients are reported for the two groups: 6-year-old and 10-year-old children.

According to our hypothesis, positive relation between metalinguistic awareness and definitional skills, as measured by both scales, emerged in both groups.

Metalinguistic Test – 6 years old

Metalinguistic Test – 10 years old

Formal Definitional Scale .435* .545*

Content Definitional Scale .554** .585*

* p<.01 (two-tailed test), ** p<.001 (two-tailed test).

Table 5. Correlation coefficients between the two Definitional Scales and the Metalinguistic Test

Relations between school success, definitional skills and metalin-guistic awareness As regards the relation between definitional skills, metalinguistic ability

and school achievement, further Correlational Analyses were performed sep-arately for the two age-levels. Pearson’s correlation coefficients for the two groups are presented in Table 6 and 7, respectively.

Formal Definitional Scale

Content Definitional Scale Metalinguistic Test

Scho

ol m

arks Marks in Italian

language .233 .244 .339*

Marks in mathematics .258 .351* .249

Teac

hers

’ Rat

ings

Lexical Knowledge .434** .541*** .459***

Syntactic Competence .289* .397** .377**

Communicative Com-petence .159 .294* .346*

Role Taking .275 .375** .326*

* p<.05 (two-tailed test), ** p<.01 (two-tailed test), *** p<.001 (two-tailed test).

Table 6. Correlations between School Success, Definitional Scales and the Metalinguistic Test for 6-year-old children


257

Formal Definitional Scale

Content Definitional Scale Metalinguistic test

Scho

ol m

arks Marks in Italian

language .502*** .627*** .576**

Marks in mathematics .335* .474** .336*

Teac

hers

’ Rat

ings

Lexical Knowledge .440** .532*** .547***

Syntactic Competence .368* .493*** .545***

Communicative Competence .422** .432** .702***

Role Taking .292 .389* .555***

* p<.05 (two-tailed test), ** p<.01 (two-tailed test), *** p<.001 (two-tailed test)

Table 7. Correlations between School Success, Definitional Scales and the Metalinguistic Test for 10-year-old children

As for the relations between School Marks and the performance in the Definitional Task of 6-year-old children (Table 6), only the correlation be-tween marks in Math and the score in the Content Scale was significant.

On the other hand, School Marks in Language are positively correlated with the performance in the Metalinguistic test.

As regards Teachers’ Ratings data show that for 6-year-old children all the four dimensions (Lexical Knowledge, Syntactic Competence, Commu-nicative Competence, Role Taking) positively correlate with the communica-tive efficacy of pupils’ definitions (the Content Scale) and with their Meta-linguistic performance. As about the Formal Scale, instead, at this lower level of age it correlates only with Lexical Knowledge and Syntactic Com-petence. For 10-year-old children (Table 7), all the variables positively corre-late with each other, in that both scores in Formal and Content definitional scales correlate with School Marks and Teachers’ Ratings. At this level of age, moreover, children’s Metalinguistic performance correlates with School Marks – in both language and maths -, as well as with Teachers’ Ratings in all the four dimensions.


258

DISCUSSION

The study presented in this paper considered the complex relations be-tween definitional skills, metalinguistic awareness and school achievement of children, at two very different levels of schooling: the beginning (first grade) and the end of Italian elementary school (fifth grade).

First of all, the first analyses showed a clear change both in definitional ability and in metalinguistic awareness for 10-year-old pupils if compared with the younger ones. At the end of primary school both abilities seem well developed and, in particular, the quality of children’s definitions is very close to that of adults’, that is the conventional Aristotelian format. Generally speaking, therefore, our data are consistent with many other studies which have shown that as a function of age definitions tend to conform to adult-like formal definitions (Johnson & Anglin, 1995; Snow, 1990).

As regards the main aim of the present paper, which was to analyze the possible relations between these two abilities and school achievement, some important results emerged.

In the first place, our data confirmed that metalinguistic ability can be considered a fundamental component in the development of good defini-tions. So, our study can be considered a first attempt to empirically demon-strate – by means of an independent metalinguistic task – the metalinguistic nature of formal definitions, even at the first phases of their acquisition (see MacGhee-Bidlack, 1991; Snow, 1990; Watson, 1985).

Second, our data evidenced the existence of a complex relation between definitional skills and school success, which changes as a function of age. This is probably due to the fact that, differently from the beginning of el-ementary school, at the last age-level, school tasks are complex enough to elicitate their relations with definitional ability and metalinguistic awareness (Cornoldi, 1995; Snow, 1990; Snow et al., 1989). In particular, as regards the relations between definitional skills and school success, it emerged that the formal quality of first-graders’ definitions is positively related with teachers’ ratings on lexical knowledge and syntactic competence (in other words the “core” aspects of definitions, that is knowledge of the definiendum and of the formal structure of the definiens), whereas in older children also communica-tive competence is related with the formal quality of definitions. Lack of relation between definitional skills and role taking ability is probably due to


259

the fact that role taking still remains a more social-interactive competence not strictly connected with linguistic-communicative performances.

Instead, data about the relations between the definitional task and school marks in 6-year-old pupils and, specifically, the existence of a significant cor-relation only between marks in mathematics and the performance in the Con-tent Scale, are more difficult to be interpreted.

Generally speaking, our study confirmed the positive relation between the development of good formal definitions and school achievement already evidenced in literature, even though the direction of this relation can not be inferred from our data (Snow et al., 1989).

As regards the relations between metalinguistic awareness and school suc-cess, it is interesting to note that in 6-year-old children metalinguistic ability is significantly correlated only with marks in Language and not with marks in Mathematics. In this case, metalinguistic competence seems to be confined to a “specific domain”, such as language learning, that is it seems to be a direct reflection of language use and language learning. On the contrary, at the end of primary school, metalinguistic awareness appears to be a more general competence, positively connected across different disciplinary domains, such as language acquisition and mathematical reasoning, and dealing with more general abstract-symbolic fields.

Actually, contents and tasks proposed to 10-year-old children are charac-terized by a higher level of symbolic formalization and likely require a meta-linguistic-metacognitive ability in problem-solving, in abstract reasoning and in manipulating symbols and rules (Davidson, Deuser & Sternberg, 1994; Schoenfeld, 1985). Therefore, at this level of age, metalinguistic ability nec-essarily intervenes in understanding mathematics problems, which requires good comprehension of written text, strategic control and monitoring of their variables.

In conclusion, our study confirmed the positive relation between school experience and ability to define words. On the one hand, individuals gradu-ally learn both the correct formal structure of the adult-like definitions and its content constraint (that is selection of the most relevant information) through examples of “good definitions” they can experience at school, i.e. in teachers’ speeches and in text-books, even when definitions are not the real issue of teaching (Snow, 1990). On the other hand, the attainment of a good level of definitional skills and the development of their different components


260

represent a valid means of comprehension and learning for pupils. In other words, definitions can be used to stipulate meanings, especially in particu-lar learning fields where discourses refer to highly abstract concepts, such as scientific-mathematical subject-matters (“A circle is the locus of points equidistant from a given point”). Due to this relation with school success, as well as with metalinguistic awareness, the analysis of children’s level in the development of definitional skills, therefore, may be highly informative for teachers and school psychologists, and it may represent a reliable measure for the assessment of pupils’ cognitive and communicative abilities.

REFERENCES

Anderson, R., & Freebody, P. (1983). Reading comprehension and the assess-ment and acquisition of word knowledge. In B. Hutson (Ed.), Advances in reading/language research: A research annual (pp. 231-256), Greenwich, CT: JAI Press

Benelli, B. (1988). If it is a dog, can it be an animal? The role of metalinguis-tic knowledge in the acquisition of linguistic superordination. Journal of Psycholinguistic Research, 17(3), 227-243

Benelli, B., Arcuri, L., & Marchesini, G. (1988). Cognitive and linguistic fac-tors in the development of word definitions. Journal of Child Language, 15, 619-635

Benelli, B., Belacchi, C. & Lucangeli, D. (2004). The ambergris is not a fossil stone: the role of form and content in the development of definitional skills. Unpublished Manuscript

Borkowski, J. G., Ryan E. B., Kurtz B. E., & Reid, M. K. (1983). Metamemory and metalinguistic development: Correlates of children’s intelligence and achievement. Bulletin of the Psychonomic Society, 21(5), 393-396

Bruner, J. (1986). Actual minds, possible worlds. Cambridge, London: Harvard University Press

Carlisle, J. F., Beeman, M., Davis, L. H., & Spharim, G. (1999). Relationship of metalinguistic capabilities and reading achievement for children who are becoming bilingual. Applied Psycholinguistic, 20, 459-478

Chall, J. (1987). Two vocabularies for reading: Recognition and meaning. In M. G. McKeown & M. E. Curtis (Eds.), The nature of vocabulary acquisition


261

(pp. 7-18). Hillsdale, NJ: Lawrence Erlbaum AssociatesCornoldi, C. (1995). Metacognizione e apprendimento. Bologna: Il MulinoDavidson, J. E., Deuser, R., Sternberg, R. J. (1994). The role of metacognition

in problem solving. In J. Metcalf & R. Shimamura (Eds.), Metacognition : knowing about knowing. Cambridge, London: The MIT press, 207-227

Davidson, R. G., Kline, S. B. & Snow, C. E. (1986). Definitions and definite noun frases: Indicators of children decontextualized language skills. Jour-nal of Research in Childhood Education, 1, 37-48

Dreher, M. J., & Zenge, S. (1990). Using metalinguistic awareness in first grade to predict reading achievement in third and fifth grades. Journal of Educational Research, 84, 13-21

Ferreira, F., & Morrison, F. J. (1994). Children’s metalinguistic knowledge of syntactic constituents: effects of age and schooling. Developmental Psychol-ogy, 30(5), 663-678

Francis, N. (1999). Bilingualism, writing, and metalinguistic awareness: Oral-literate interactions between first and second languages. Applied Psycholin-guistics, 20, 533-561

Grice, H. P. (1975). “Logic and conversation”. In P. Cole, F.L. Morgan (eds), Speech acts: syntax and semantics, vol. 3. New York: Academic Press

Guiterrez-Clellen, V. F., & De Curtis, L. (1999). Word definition skills in Spanish-speaking children with language impairment. Communication Disor-ders Quarterly, 21(1), 23-31

Hakes, D. T. (1980). The development of metalinguistic abilities in children. New York: Springer-Verlag

Johnson, C.J. & Anglin, J.M. (1995). Qualitative developments in the content and form of children’s definitions. Journal of Speech and Hearing Research, 38, 612-629

Kikas, E. (1993). The development of word definitions in children. Journal of Russian and East European Psychology, 31, 40-54

Kurland, B. F., & Snow, C. (1997). Longitudinal measurement of growth in definitional skill. Journal of Child Language, 24, 603-626

Litowitz, B. (1977). Learning to make definitions. Journal of Child Language, 4, 289-304

McGhee-Bidlack, B. (1991). The development of noun definitions: a metalin-guistic analysis. Journal of Child Language, 18, 417-434

Menyuk, P. (1984). Language development and reading. In J. Flood (Ed.), Un-


262

derstanding reading comprehension (pp. 101-122). Newark, DE: International Reading Association

Nagy, W., & Herman, P. (1987). Breadth and depth of vocabulary knowledge: Implications for acquisition and instruction. In M. McKeown & M. Curtis (Eds.), The nature of vocabulary acquisition (pp. 9-35). Hillsdale, NJ: Erlbaum

Nelson, K. (1996). Language in cognitive development: emergence of the mediated mind. Cambridge: Cambridge University Press

Nippold, M. A. (1995). School-age children and adolescents: Norms for word definitions. Language, Speech, and Hearing Services in Schools, 26, 320-325

Olson, D. R. (1991). Literacy as metalinguistic activity. In D. R. Olson, N. Tor-rance (Eds.), Literacy and orality. Cambridge: Cambridge University Press

Olson, D. R., & Astington, J. W. (1993). Thinking about thinking: learning how to take statements and hold beliefs. Educational Psychologist, 28, 7-23

Rey-Debove, J. (1971). Etude linguistique et sémiotique des dictionnaires français con-temporains. Paris: The Hague: Mouton

Schneider, W., & Naeslund, J. C. (1993). The impact of early metalinguistic competencies and memory capacity on reading and spelling in elementary school: results of the Munich Study on the Genesis of Individual Compe-tencies. European Journal of Psychology of Education, 8(3), 273-287

Schoenfeld, A. H. (1985). Metacognitive and epistemological issues in math-ematical understanding. In E. A. Silver (Ed), Teaching and learning mathemati-cal problem solving. Hillsdale, NJ: Erlbaum, 361-379

Sinclair, A. (1986). Metalinguistic knowledge and language development. In I. Kurkz, G. W. Shugar, J. H. Danks (Eds.), Knowledge and language. Amster-dam: North-Hollandù

Smith, C. L., & Tager-Flusberg, H. (1982). Metalinguistic awareness and lan-guage development. Journal of Experimental Child Psychology, 34, 449-468

Snow, C.E. (1990). The development of definitional skill. Journal of Child Lan-guage, 17, 697-710

Snow, C. E., Cancino, H., Gonzalez, P., & Shriberg E. (1989). Giving formal definitions: An oral language correlate of school literacy. In D. Bloome (Ed.), Classroom and literacy. Norwood, NJ: Ablex

Snow, C. E., Cancino, H., De Temple, J., & Schley, S. (1991). Giving formal definitions: a linguistic or metalinguistic skill? In E. Bialystok, Language processing in bilingual children. New York: Cambridge University Press

Storck, P. A., & Looft, W. R. (1973). Qualitative analysis of vocabulary re-


263

sponses from persons aged six to sixty-six plus. Journal of Educational Psy-chology, 65, 192-197

Tulviste, T. (1993). The function of differentiating between a word and its referent. Journal of Russian and East European Psychology, 31(2), 27-39

Watson, R. (1985). Towards a theory of definition. Journal of Child Language, 24, 535-565

Watson, R. (1995). Relevance and definition. Journal of Child Language, 22(1), 211-222

Watson, R., e Olson, D. R. (1987). From meaning to definition: A literate bias on the structure of word meaning. In R. Horowitz e J. S. Samuels (Eds.), Comprehending oral and written language. San Diego: Academic Press

Wehren, A., DeLisi, R., & Arnold, M. (1981). The development of noun defi-nition. Journal of Child Language, 8, 165-175

Williams, R. H. (1992). Tukey-like pairwise comparison among proportions. Educational and Psychological Measurement, 52, 913-914

ABSTRACT

Definitional skills, as a measure of decontextualized language abilities, are closely related to school success, in general, and the development of literacy, in particular (Snow et al., 1989).

The purpose of the present study is to investigate the relations between definitional skills, metalinguistic awareness and school achievement in Italian children at the beginning (first grade) and the end (fifth grade) of primary school. Participants were asked to define a list of 24 words (Nouns, Verbs and Adjectives) and their answers were scored for both form and content; the level of their metalinguistic ability was measured by means of a questionnaire concerning different components of metalinguistic competence.

School achievement was assessed through pupils’ school marks; further-more, teachers’ ratings of children’s linguistic-communicative ability were collected and correlated with definitional skills. Results clearly show positive relations between definitional skills, metalinguistic awareness and school suc-cess at both levels of age.


264

Acknowledgements: We would like to thank all teachers and pupils who participated in the study.

Key words: definitional skills; metalinguistic awareness; school success

Received September 29 2003; accepted November 18 2004


265

APPENDIX

The Metalinguistic Test

Question 1. Do you know what a word is? Give me an example.

Score:1 = no answer or answer without meaning (ex. “I don’t know”);2 = wrong answer with or without justification (ex. “it is the alphabet”);3 = functional right answer (ex. “for speaking”);4 = linguistic right answer (ex. “it is a set of letters”).

Question 2. Is “ball” a word or not? Why? If not, what is it? Is “when” a word or not? Why? If not, what is it? Is “xbl” a word or not? Why? If not, what is it?

Score (ex. for “ball”): 1 = no answer or answer without meaning (ex. “I don’t know”);2 = negative answer (no) with or without justification (ex. “no, because I

don’t like to shoot it”);3 = positive answer (yes) without justification or with referential-semantic

justification (ex. “yes, because it is a game”);4 = positive answer (yes) with linguistic justification (ex. “yes, because it in-

cludes letters”).

Question 3. Do you know what a definition is? What does it mean “to de-fine a word”?

Score:1 = no answer or answer without meaning (ex. “I don’t know”);2 = wrong answer (ex. “that you must finish a word”);3 = functional answer (ex. “it means to write”);4 = definitional answer (ex. “that you explain how a thing is”).

Question 4. These are three different ways I could use to define, to ex-plain what a word means. Tell me what is the best, according


266

to you.

Orange: a) It is done of segment and you can squeeze it; b) It is a thing done of segment and you can squeeze it; c) It is a fruit done of segment and you can squeeze it.Why the form you chose is better than others?

Football player: a) He is someone who kicks the ball at the stadium; b) He is a person who kicks the ball at the stadium; c) He kicks the ball at the stadium.

Why the form you chose is better than others?

Table: a) It is a piece of furniture you can use to eat on, to write and to work; b) You can use to eat on, to write and to work; c) It is a thing you can use to eat on, to write and to work.Why the form you chose is better than others?

Score: 1 = no answer or answer without meaning (ex. “I don’t know”);2 = descriptive answer (a);3 = answer with the generic superordinate (b);4 = answer with the specific superordinate (c).Why the form you chose is better than others?

Score:1 = no answer or answer without meaning (ex. “I don’t know”);2 = tautological answer (ex. “because it is an orange”);3 = referential-semantic answer (ex. “because you can do the orange juice”);4 = linguistic answer (ex. “because you can understand better”).

Question 5. Could a “cat” be called an “animal”? Why?Could a “cat” be called a “kitten”? Why?Could a “cat” be called an “horse”? Why?

Could a “car” be called a “machine”? Why?


267

Could a “car” be called a “bicycle”? Why?Could a “car” be called a “means of transport”? Why?

Could a “doctor” be called an “baker”? Why?Could a “doctor” be called a “man”? Why?Could a “doctor” be called a “physician”? Why?

Score: 1 = no answer or answer without meaning (ex. “I don’t know”);2 = negative answer (no) with or without justification;3 = positive answer (yes) without justification or with justification which re-

fers to the instrumental use of the word (ex. “yes, because in that way the cat comes”) – observation (ex. “yes, because it is an animal”);

4 = positive answer with linguistic justification (ex. “yes, because it can be is name”).

Question 6. Where do names of objects come from?

Score:1 = no answer or answer without meaning (ex. “I don’t know”);2 = natural origin (ex. “names are inside things”);3 = transcendent origin (ex. “God decided names”);4 = cultural origin (ex. “people gave names to things”).

Documents

Children’s definitional skills and their relations with metalinguistic awareness and school achievement