Collaborative inquiry and the construction of explanations in the learning of science

Learning and Instruction 12 (2002) 189–212www.elsevier.com/locate/learninstruc

Collaborative inquiry and the construction ofexplanations in the learning of science

S. Kaartinen*, K. KumpulainenDepartment of Educational Sciences and Teacher Education, University of Oulu, P.O. Box 2000,

FIN-90014 Oulu, Finland

Abstract

This study investigates the construction of explanations in a social science-learning situation,in which 18 university students participated. The science-learning task which implicitly modelleddesign principles for science instruction in school contexts was derived from kitchen chemistrywhere the students investigated in small groups the nature of five solid samples of differentcooking ingredients. The instructional goal of the learning situation was to develop the students’conceptions of solubility, the activity itself involved collaborative inquiry and experimentation.The study follows a three-step research design: pre-test, intervention and post-test, in order tohighlight the students’ explanations around the concept of solubility and their elaboration in socialactivity. A specific discourse analysis method was developed for the study, to investigate themechanisms of explanation-building in small-group discourse. The data for the study were col-lected by means of videotapes, direct observations, transcriptions and questionnaires clarifyingthe students’ explanations for dissolving. The study introduces an analytic tool for untangling theprocesses of explanation-building in collaborative inquiry which takes a synchronous and dia-chronic approach to communicative and cognitive processes of student discourse. The analysishighlights the reciprocal relationship between the nature of explanations and the students’ com-municative processes in the evolving discourse. The data analysis shows that the negotiationprocesses around the concept of solubility consisted of diverse interpretations, varying from infor-mal explanations to formal explanations, and from descriptive reasoning to causal reasoning. Theresults of the students’ pre- and post-tests indicate that the social science-learning situation pro-vided the students with opportunities to elaborate their explanations for dissolving, and reflectingpractical, theoretical and applied understanding. 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Science-learning situation; Explanation-building; Analytic tool

* Corresponding author. Tel.:+358-8-5533713; fax:+358-8-5533600.E-mail address: [email protected] (S. Kaartinen).

0959-4752/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved.PII: S0959 -4752(01 )00004-4

190 S. Kaartinen, K. Kumpulainen / Learning and Instruction 12 (2002) 189–212

1. Introduction

Although classroom discourse in science education is increasingly becoming aprominent area of research in education (Cobb & Bowers, 1999; Howe, Tolmie,Greer, & Mackenzie, 1995; Kelly & Chen, 1999; Lemke, 1990; Roth & Lucas, 1997),little is still known on explanation-building processes in student-centred small-grouplearning situations. In particular, there is little evidence about the nature of discoursein science-learning contexts dependent on argumentation and persuasion, and wherestudents have the possibility to bring in their informal and formal explanations forjoint negotiation and justification. Further, we lack analytical tools to highlight expla-nation-building processes in students’ discourse while working on small-groupscience tasks (Kelly & Chen, 1999).

In this article, we examine discourse processes and explanation-building in a uni-versity science class working on a group investigation task around the concept ofsolubility. The study is shaped by the sociocultural views of learning (Cole, 1996;Vygotsky 1962, 1978; Wells, 1999; Wertsch, 1991). According to this line of think-ing, science learning can be viewed as a participatory, interactional process whichincludes the negotiation of the cultural practices of scientific communities. Thesecultural practices reflected in the discourse of science classrooms include con-structing explanations, defending and challenging claims, interpreting evidence,using and developing models, transforming observations into findings, and arguingtheories (Coleman, 1998; Herrenkohl & Guerra, 1998; Hogan, Nastasi, & Pres-sley, 2000).

In order to understand the nature of explanation-building during students’ groupinvestigation, a specific discourse analysis method was designed for the presentstudy. The analytic tool takes a synchronic and diachronic perspective on discourse.The synchronic perspective is achieved by concentrating on four parallel analyticframes, i.e. discourse moves, logical processes, nature of explanation and cognitivestrategies, whereas the diachronic perspective is reflected in the microanalysis of thestudents’ discourse which evolves on a moment-by-moment basis.

Later in this article we shall shortly review studies investigating students’ expla-nations in the learning of science. Firstly, in the theoretical discussion attention isgiven to the sociocultural views on learning and their application to science instruc-tion. Secondly, the discourse analysis method developed for this study will be high-lighted with a case study showing the application and potential of the method to theinvestigation of students’ discourse processes and explanation-building in socialscience learning. These analysis are aimed at providing tentative evidence on thenature of students’ learning in the science task.

2. Theoretical framework

The nature and development of students’ explanations in science learning havebeen studied using the conceptual change approach, which has its roots in two rela-tively independent research traditions, namely the cognitive developmental tradition

191S. Kaartinen, K. Kumpulainen / Learning and Instruction 12 (2002) 189–212

and the science education tradition (Vosniadou, 1999). The science educationapproach to conceptual change suggests that in science-learning situations, studentsbring in their alternative frameworks and misconceptions, which are robust and dif-ficult to change through teaching (Viennot, 1979; Driver & Easley, 1978). Develop-mental research on conceptual change concentrates on investigating learners’ concep-tual knowledge and its relationship to scientific knowledge particularly in their earlyyears (Carey, 1985). There has also been a growing interest in understanding howconceptual structures change in the process of development and with the acquisitionof expertise (Vosniadou, 1994; Chi, Slotta, & de Leeuw, 1994).

Besides the conceptual change approaches, the sociocultural perspective hasstarted to provide alternative insights to science learning and instruction (Cobb &Bowers, 1999; Kelly & Green, 1998; Saljo, 1999). The social views of learning anddevelopment have their roots in the work of Mead (1934), Vygotsky (1962, 1978)and their followers (Bruner, 1996; Cole, 1996; Rogoff, 1990; Wertsch, 1991).According to this school of thought, the engagement in cultural activities is consti-tuted of dialogic, changing and transformative practices. From this perspective, learn-ing is seen as an open-ended process with the possibility of diverse ways of acting,feeling and thinking. In viewing learning as a process of participation in culturallyorganised activities, the sociocultural views of learning emphasise the role of langu-age and other semiotic tools in dialogic meaning-making (Wells, 1999). In scienceeducation, the cultural artefacts, such as oral and written language and laboratoryequipment, are seen as providing a shared semiotic system for social interaction andmodes of thinking (Kelly & Chen, 1999; Saljo, 1999; Vygotsky, 1962). The semiotictools negotiated and re-negotiated in collective activity create the grounds for mean-ing-making (Wertsch, 1991).

In summary, in the conceptual change approaches learning is often viewed as anacquisition and accumulation process in which the learner gradually gains ownershipof the concept or knowledge in question (Sfard, 1998). For the sociocultural frame-work, learning is regarded as a participatory process in which the learner graduallybecomes an active member in a cultural community by learning its discourse prac-tices, norms and ways of thinking. From this perspective, knowing refers to belong-ing, participating and communicating (Wenger, 1998) instead of having or pos-sessing, as usually suggested by cognitively oriented approaches to learning anddevelopment (Schnotz, Vosniadou, & Carretero, 1999).

The social views of learning challenge pedagogical and methodologicalapproaches to design, and examine science-learning situations. An important peda-gogical question to be considered is whether the instructional setting engages thestudents in the authentic practice of science, i.e. performing and interpreting experi-ments for the specific purpose of verifying or revising theories and explanations, andcommunicating and negotiating them among the group members (Kelly & Chen,1999; Schauble, Glaser, Duschl, Schulze, & John, 1995). Methodologically, there isa need to develop analytical tools which offer a dynamic and process-orientedaccount on collaborative science-learning, and which untangle the interplay betweenthe communicative and cognitive elements of discourse processes. Analyses of this


nature are likely to increase the current understanding of collaborative inquiry in thelearning of science (Kumpulainen, 1996; Kumpulainen & Mutanen, 1999).

2.1. Collaborative inquiry in classroom practice

Experimenting plays a crucial role in typical school-based science-learning situ-ations. However, usually its role is to verify the theory that has been presented tothe students in instruction. This pedagogical practice does not necessarily guaranteethat the students’ conceptual understandings converge with the scientific knowledgeembedded in the instructional setting (Schauble et al., 1995). The reason for thismay lie in the fact that this kind of learning situation contrasts with real-life learning,where the problem is posed at the beginning of the problem-solving process (Watts,1991). The acquisition-oriented view of science education stresses on searching exactanswers and controlled procedures in experimenting (Sfard, 1998). The applicationof this view to science education is often reflected in authorised classroom practiceswhere the teacher is “an authority of science” for his/her students (Crawford,Kelly, & Brown, 2000; Russell, 1983). How will the situation be changed if thestarting point is a learner and his or her special interest? For the student the practiceof knowledge construction might not be just to predict, explain or control the regu-larities of nature. Instead, science learning could consist of the kind of possibilitiesfor thought and action opened up, in a sense, by immeasurable reality (von Gla-sersfeld, 1989). In pedagogical practice this could be realised in learning activitieswhere students approach the theme of inquiry by posing meaningful, self-initiatedquestions and elaborate them in collective activity. During the ongoing dialogue thecommunity may question, re-interpret, elaborate and verify explanations and relatethem to existing theories.

The instructional design of the present study is aimed at providing enabling con-ditions for collaborative inquiry in learning chemistry. In this study, the connectionbetween experimental work and explanation-building elements in the science-learn-ing situation were based on the assumption that human activity is conceived to besimultaneously mental and practical (Saljo, 1999). The pedagogical principles ofthe learning situation emphasised student-initiation and interest, problem posing andsolving, interpretation and speculation, and distributed expertise. In this learning con-text, the teacher was seen as a participant of collaborative inquiry who contributedto the discourse with his or her expertise as a scientist, educator and human being.To summarise, the key elements of collaborative inquiry in this study were exper-imentation, social negotiation and explanation-building.

3. The study

3.1. Research goals

The goal of this study was to investigate explanation-building processes in a socialscience learning situation with adult students. An analytic tool for highlighting the


mechanisms of explanation-building in social interaction was developed. The appli-cation and potential of this method is demonstrated with a case study discussed inthis article. The case study consists of pre- and post-test analysis of students’ expla-nations for dissolving and a microanalysis of the discourse of one student groupworking on the science-learning task.

The specific research goals for this study are:

� To develop an appropriate analytic tool to investigate explanation-building pro-cesses in social science learning.

� To investigate explanation-building processes in collaborative inquiry.� To investigate the connection of students’ personal explanations to those expla-

nations negotiated in collaborative inquiry.

3.2. Subjects

Eighteen university students participated in this study along with their scienceteacher at the Department of Teacher Education, Oulu University, Finland. The sub-jects were second-year students attending a compulsory course on chemistry teach-ing. The average age of the students, 8 males and 10 females, was 23.6 years. Thestudents’ skills in chemistry represented an average level as assessed by the univer-sity lecturer responsible for the course.

3.3. Description of the learning situation

The context of the learning situation in which the students worked was derivedfrom kitchen chemistry. The instructional goal of the activity was to develop thestudents’ conceptions of solubility, the activity itself involved collaborative inquiryand experimentation. The concept of solubility is central to the understanding of thechemical properties of solutions and, consequently, it played a central part in thewhole chemistry course the students took part in.

In the learning situation, the students worked in self-selected small groups. Theaverage size of the mixed-gender groups was four students. The whole class workedsimultaneously in the same classroom, carrying out their research designs for ident-ifying samples. The students’ research practices consisted of making hypotheses,observations and conclusions.

In the activity, the students were given five solid samples of different cookingingredients to be analysed. These five samples were salt, sugar, potato flour, bakingsoda and one mixture consisting of wheat flour and baking soda. All were similarin appearance, being white powders. In order to identify the samples it was necessaryto apply strategies for scientific investigation. These included developing a reliablemethod to solve the problem in question. Student-initiated experimental work anddiscussions arising from it created a platform for the students’ negotiation processesabout the meaning of solubility. The educational goal of the science course as a


whole was to develop the students’ conceptions of learning and instruction in scienceas well as to elaborate their conceptual understanding.

3.4. Methods and procedures

The data collected for the study consist of one 4-h instructional unit, realised assmall-group work activity. These data are part of a larger research project investigat-ing science learning and instruction in teacher education. The primary data comprisedvideotaped and transcribed episodes of small-group interaction occurring in thescience class. The video data cover the social interaction of the whole science class.

In order to evaluate the students’ conceptions about the meaning of dissolving,the students were administered a pre- and post-test which they completed individuallybefore and after the instructional unit. The pre-test was administered on the sameday that the unit started, whereas the post-test was administered two weeks after theinstructional unit. The pre- and post-tests consisted of one open-ended question ask-ing the students to clarify the processes involved in dissolving. The specific questionwas “Describe the processes involved in dissolving”?1 The students answered thequestion in writing in their science classroom.

3.5. Logic-of-inquiry for investigating discourse processes and explanation-building

The logic-of-inquiry applied in the present study to investigate discourse processesand explanation-building in social science learning is influenced by the work ofBarnes and Todd (1995) and Kumpulainen and Mutanen (1999) as well as by interac-tional ethnographers (Gee & Green, 1998; Kelly & Chen, 1999). Domain-specificknowledge about science and science education has also played an important rolein guiding the development of the analysis method and in coding the interaction data.

The analysis developed for this study focuses on four parallel analytic frames, i.e.discourse moves, logical processes, nature of explanation and cognitive strategies.The analysis of discourse moves and logical processes is aimed at providing insightsinto the nature of reasoning and explanation-building from the viewpoint of partici-pation in social activity. The analysis of the nature of explanation and cognitivestrategies is aimed at unravelling conceptual and procedural elements in reasoningand explanation-building. Whereas the main analytical frames of the coding schemeare derived from the existing literature, the specific analytic categories are based onthe interaction data of this study. Consequently, the categories have been contextuallydefined on a post hoc basis.

In order to develop the analysis to investigate collaborative inquiry and expla-nation-building in the science-learning situation, the interaction data collected wereanalysed in several phases. In the first phase, the video material capturing the social

1 A formal description of dissolving: A real solution is formed when a solute is dissociated in solventinto molecules or ions (from the teacher’s notes).


activity of the science class was closely examined, first independently and then col-laboratively by two researchers. The analysis of the videotapes was supported bythe researcher’s field notes from the science course. Next, the discourse occurringin the science class was transcribed. The final construction of analytic categorieswas based on the transcriptions, supported by the video and observational data.Although the code names of the analytic categories grounded in the interaction datarepresent context-free codes, the meaning behind the code names has been contex-tually defined to represent the interaction data of this study. The analytic categorieswhich emerged from the interaction data are characterised in more detail in Section4. Table 1 summarises the analytic frames and categories of the analysis method.

The application of the coding scheme is realised through the microanalysis ofevolving small-group interaction by focusing on each conversational turn, usingmutually exclusive and exhaustive categories. The reliability of the coding of thestudents’ discourse has been checked by two independent researchers (one scienceeducator and one researcher in educational sciences) who have analysed the data.The inter-rater agreement between the coders was 83.8%. Diverse opinions havebeen negotiated to establish a joint agreement. The construction of a joint agreementhas been guided by domain-specific knowledge of chemistry. Owing to the interpret-ative and complex nature of the analysis this procedure was found most appropriateto the rationale of this study.

3.5.1. Discourse movesThe categories characterising discourse moves are adapted from the framework

developed by Sinclair and Coulthard (1975). The discourse moves in the method areidentified from each conversational turn on the basis of their relationship to logicalprocesses, cognitive strategies and the nature of explanations. The analysis of dis-course moves highlights the nature of conversational exchanges between students intheir social interaction and, consequently, sheds light onto the participatory roles ofgroup members in social science learning. Moreover, the analysis of discourse movessupports content analysis by highlighting thematic patterns emerging in joint problemsolving. Discourse moves identified in the analysis framework are initiating, continu-

Table 1The analytic method for analysing social interaction

Discourse moves Logical processes Nature of explanation Cognitive strategies

� Initiating � Proposes a cause � Formal explanation � Constructing a� Continuing � Proposes a result � Causal explanation question� Extending � Advances evidence � Descriptive � Raising a new� Referring back � Suggests a method explanation question� Agreeing/disagreeing � Evaluates � Everyday � Using evidence� Replying � Contradicts explanation � Applying a principle� Commenting to a case (modelling)� Concluding � Using everyday

knowledge


ing, extending, referring back, agreeing/disagreeing, replying, commenting and con-cluding. Turns coded as initiation moves begin new thematic interaction episodesbased on close domain-specific analysis of discourse. Continuing moves are definedas reflecting the students’ joint interpretation of the situation in which the studentscontinue to elaborate either their own or their peers’ reasoning. Extending moves areseen as bringing in new perspectives which expand joint explanation-building underthe same theme. Referring back moves refer back to the ideas which have alreadyemerged in the flow of discourse. Agreeing/disagreeing moves signal the acceptanceor rejection of ideas and explanations proposed in the previous conversational turns.Replying moves are defined as responses to explicit questions. Commenting movesare statements uttered in the course of discourse to give personal remarks or evalu-ations of the situation. Concluding moves draw together explanation-building pro-cesses.

3.5.2. Logical processesThe analysis of logical processes is concerned with the logical relationship

between conversational turns and how they give rise to explanation-building in socialinteraction (Barnes & Todd, 1995). The analysis of logical processes gives meaningto discourse moves and supports domain-specific interaction analyses. In addition,it provides further evidence on the nature of the students’ participatory processes insmall group work. Logical processes categories identified in the interaction analysisframework are proposes a cause, proposes a result, advances evidence, suggests amethod, evaluates and contradicts. The proposes a cause category represents twotypes of reasoning: describing causes as processes (i.e. characterising the process ofsolute–solvent interaction) or describing causes as factors (i.e. defining the reasonfor a solute–solvent interaction). The proposes a result category conceptualises theoutcome of cause–effect reasoning (e.g. defining the product of solute–solventinteraction). The advances evidence category shows that conceptualising is deepenedon the basis of formal or informal reasoning. The suggests a method category high-lights procedural strategies in the students’ inquiry. The evaluates category showscritical assessment of joint problem-solving. The contradicts category brings in dis-crepancies in the students’ reasoning and problem solving. In the present study notall conversational turns were found to reflect logical processing and consequentlynot all were coded.

3.5.3. Nature of explanationThe specific categories describing the nature of explanations are formal expla-

nation, causal explanation, descriptive explanation and everyday explanation.With regard to the nature of explanations, the category formal explanation is

defined as reflecting the formal language and procedures of chemistry. Explanationscoded in this category do not contradict those uttered by experts of the scientificdomain. The causal explanation category includes explaining the cause and/or resultof dissolving with the help of informal language. Explanations coded in this categoryare more informal in nature than formal explanations and they are not necessarilycongruent with those of experts. The descriptive explanation category characterises


the process of dissolving but does not explain specifically the causal relationshipsof solute and solvent. The category of everyday explanation reflects the creation ofsituational meanings derived from informal contexts. Table 2 shows the categoriesof the nature of the students’ explanations as identified in the discourse data.

3.5.4. Cognitive strategiesThe categories describing the nature of cognitive strategies in joint problem solv-

ing include constructing a question, raising a new question, using evidence, applyinga principle to a case and using everyday knowledge. The constructing a questioncategory refers to a situation in which a problem is framed. Raising a new questiondemonstrates the emergence of sub-questions for a problem. The using evidencecategory identifies a situation in which the student is reasoning through experimen-tation and conceptualising. The category applying a principle to a case reflects amodelling activity in which scientific knowledge is applied to a specific case. Usingeveryday knowledge refers to a situation where reasoning is based on informal, every-day experiences.

3.6. The evaluation of the pre- and post-tests

The pre- and post-tests were administered to the students before and after thesmall-group science-learning situation to investigate the possible changes in the nat-ure of the students’ explanations for dissolving. The students’ written descriptionswere analysed qualitatively with the help of descriptive categories arising from the

Table 2Categories describing the nature of explanations in the students’ discourse

Nature of explanations Description Example

Formal explanation Reflects the formal language and In dissolving it seems to me thatprocedures of chemistry whilst the crystal structure gets smallerconceptualising solute-solvent in salt so that the sodium…interaction

Causal explanation Includes informal reasoning There should be a formula…if itwhilst conceptualising solute- didn’t dissolve, then there shouldsolvent interaction still be those crystals here

Descriptive explanation Describes solute-solvent I think it’s like this…originallyinteraction based on there was sugar or its carbonexperimentation system and this water separately

and now they are bound to eachother

Everyday explanation Reflects the construction of In hospital it is possible to drinksituational meanings based on sugar-watereveryday reasoning

The salt and water havedissolved, there is no salt andwater but saltwater


data. The construction of the analytic categories were based on a close reading ofthe students’ explanations within a framework provided by domain-specific expertise.The specific categories into which the students’ responses were coded are descriptiveexplanation, practical explanation and explicatory explanation. The explicatoryexplanation has been divided into three sub-categories entitled as proposes a result,proposes a cause and provides a formal explanation. These sub-categories are relatedto each other, however there are subtle distinctions, as will be shown in Table 3below. Each student response was coded in only one of the analytic categories, asguided by domain-specific knowledge of chemistry. The analytical categories aredescribed with examples in Table 3.

The data on the students’ pre- and post-tests have been analysed by two inde-pendent researchers. The inter-rater agreement between the coders was 92.9%. Dis-agreements have been negotiated to construct a consensus.

4. Results

The results of this study are discussed in two parts. Firstly, case-based descriptionsderived from one student group will be shown to highlight the students’ discourseprocesses and explanation-building in science learning. The case-based descriptionsalso demonstrate the application of the interaction analysis method to unravel themechanisms of explanation-building in collaborative inquiry. Secondly, the resultsof the nature of the students’ explanations for dissolving in the pre- and post-test

Table 3Categories describing the nature of the students’ explanations in the pre- and post-test conditions(Question: Describe the processes involved in dissolving)

The nature of explanation Definition Examples from the data

Descriptive Describes dissolving as a The substance dissolves intoprocess. The explanation does something so that there is annot clarify causal relationships equilibrium in liquid

Practical Explains dissolving with Dissolving reminds me of dippingpractical, everyday examples a tea bag in hot water or

preparing a punchExplicatory� Proposes a result Approaches dissolving from the Two or more substances dissolve

point of view of a result by forming a mixture� Proposes a cause Explains a reason for dissolving In dissolving, for example, in

sugar, the structural parts of asolid are dissociated

� Provides a formal explanation Examines dissolving as a holistic A physical phenomenon in whichphenomenon by taking account a substance is dissociated intoof solute–solvent interaction as differenct ionswell as cause and resultrelationships


conditions will be discussed within the case group and across the students of thescience class.

4.1. A case-based description

This case-based description highlights the negotiation processes of one small groupwhile investigating the nature of five solid samples, e.g. the physical character ofthe samples, their solubility and electrical conductivity. The extract characterises thestudents’ discourse as they define dissolving. The group consists of four universitystudents who were part of a learning community of eighteen students and their tutor.The data of this group were selected for this article because it displays social expla-nation-building occupied with diverse interpretations and explanations. For this rea-son, the data of the group were also considered to provide a relevant case to demon-strate the application and potential of the interaction analysis method. Table 4 showsthe discourse data of the student group. The extract consists of 43 conversationalturns in total, from a 5-min continuous working period.

The data presented in Table 4 will be discussed here by firstly summarising thefindings from the analyses of the students’ social interaction within the group. Specialattention is paid to the identification of domain-specific thematic episodes in thestudents’ discourse. This is followed by a micro-level investigation of three interac-tion episodes of the students’ discourse.

The analysis of the students’ discourse reveals altogether six thematic episodes2

in the construction of an explanation for dissolving. The conceptual themes are dis-cussing solute–solvent interaction derived from the experimental work (Episode 1),negotiating the features of physical and chemical phenomena (Episode 2), excludingchemical change from dissolving (Episode 3), investigating properties of solutions(Episode 4), recognising sugar from salt (Episode 5), and negotiating the meaningof dissolving (Episode 6).

The analysis of discourse moves in the students’ discourse shows that the thematicepisodes usually started from the initiation move, leading to several conversationalturns which often took the form of continuing or extending. The data also show thatconceptually complex topics and explanations occurring within one thematic episode,appeared to require several conversational turns.

The analysis of the students’ procedural activity reveals diverse cognitive stra-tegies in explanation construction. In Episode 1, the student group started their workby defining the problem. In Episode 2, which was conceptually a difficult theme,the students used several strategies for problem-solving, such as modelling (i.e. con-structing a formula for dissolving), reasoning and evaluating. The richness in dis-course moves and logical processes of interaction correlates with the diverse cogni-tive strategies used in conceptualising the situation. In Episode 3, the studentsexplicated their problem-solving processes to the tutor. This interaction episode ledto Episode 4 in which the students posed a problem for investigating the properties

2 The boundary of each episode was defined on the basis of its thematic focus.


Table 4An analytical map of the discourse of one student group around the theme of solubility (the transcripthas been translated from Finnish to English)

No. Name Transcribed discourse Discourse Logical Nature of Cognitivemoves processes explanation strategies

Episode 1: dissolving — solute (sugar)–solvent (water) interaction1 Juho I think it’s like this… originally Initiating Proposes Descriptive Constructing

there was sugar or its carbon a cause explanation a questionsystem and this water separatelyand now they are bound to eachother

2 Jarmo So there was a chemical reaction Continuing Proposes Causal Raising aa result explanation new

question3 Juho Dissolved from them there has Extending Advances Everyday Using

come a third substance…this is not evidence explanation evidencepure sugar but is sugar-water

Episode 2: dissolving — physical or chemical phenomenon4 Jarmo Can we write a formula Initiating Suggests Raising a

a method newquestion

5 Paula Yes we can…what is it… water Extending Advances Formal Applying aplus sodium chloride evidence explanation principle to

a case6 Elina Sodium chloride let’s write the Continuing

formula7 Elina This is not really experimental Extending Evaluates

work8 Juho There should be a formula…if it Extending Proposes Causal Using

didn’ t dissolve…then there should a cause explanation evidencestill be those crystals here

9 Jarmo Saltwater Continuing Proposes Everyday Usinga result explanation everyday

knowledge10 Juho If there were no third Referring Proposes Causal Using

substance…then there would still back a cause explanation evidencebe those crystals

11 Elina But I think that in this task it’s not Extending Evaluatesnecessary to write a formula

12 Juho No but I thought…if there were Continuing Proposes Causal Usingcrystals…so it would be a cause explanation evidenceimpossible to give a formula

13 Elina Yes…there’d be two separate Extending Proposes Causal Applying asubstances…but now we have a a result explanation principle tosolution a case

14 Jarmo Yes Agreeing15 Juho So…here should be one form Continuing16 Elina yes…sodium chloride water- Extending Proposes Formal Applying a

solution a result explanation principle toa case

(continued on next page)


Table 4 (continued)


17 Juho Should we have a formula…does Referringthat kind of substance exist back

18 Jarmo In hospital it is possible to drink Continuing Everyday Usingsugar-water explanation everyday

knowledge

Episode 3: dissolving — physical phenomenon19 Teacher Which phase are you in right now Initiating20 Elina We are just reasoning Replying Evaluates21 Juho Probably there is a third substance Continuing Proposes

a resultTeacher Silence

22 Jarmo The salt and water have dissolved Referring Everyday Usingthere is no salt and water…but back explanation everydaysaltwater knowledge

23 Elina Yes but we could say that it is a Extending Contradicts Causal Applying asolution of sodium chloride in explanation principle towater…there doesn’ t necessarily a caseexist a third substance…it is still asolution

24 Paula Yes but if we want to dissolve Extending Advances Causal Usingmetal into some solvent we have evidence explanation evidenceto separate the metal from thesolvent…so there is something elsethan salt

25 Juho Yes Agreeing

Episode 4: evaporation as a separation method of dissolved substance26 Juho If we evaporate water Initiating Suggests Constructing

a method a question27 Elina Then the salt remains Continuing Proposes Causal Applying a

a result explanation principle toa case

28 Juho Yes Agreeing29 Jarmo Mystical…like something spiritual, Commenting

it changes from water to salt

Episode 5: recognising sugar from salt30 Elina Yes but how did we differentiate Initiating Evaluates Constructing

salt from sugar a question31 Paula From the colour Continuing Using

evidence32 Jarmo We reasoned it through our senses Extending Evaluates

(continued on next page)


Table 4 (continued)


Episode 6: the meaning of dissolving33 Juho I do not know what dissolving Initiating Evaluates

is..whether one substance givessomething to another

34 Paula You can never read about it Continuing Evaluates35 Jarmo Many times in high school, but I Commenting Evaluates

have forgotten36 Elina On the other hand, in dissolving it Extending Proposes Formal Using

seems to me that the crystal a cause explanation evidencestructure gets smaller in salt sothat the sodium

37 Juho Sodium chloride breaks down Continuing Proposes Formal Usinga cause explanation evidence

38 Elina Sodium and chloride those ions go Extending Proposes Formal Usingthere a result explanation evidence

39 Jarmo Yes the idea is like that those concluding proposes Formal Usingcrystals in a way break down a cause explanation evidence

40 Elina Yes Agreeing41 Jarmo And when water is being Continuing Proposes Formal Using

evaporated those ions join together a result explanation evidenceagain

42 Elina that is how it seems to be agreeing evaluates43 Jarmo I think it is also a chemical Continuing Evaluates

explanation on the basis offormulae

of solutions. In Episode 5 the students were engaged in evaluating their workingstrategies, whereas Episode 6 reflects evaluative, reasoning and explanatory activity.

4.1.1. Episode 1In Episode 1 the student group begins to elaborate their conceptions of solubility.

This episode suggests that the students were confused as to whether dissolving is aphysical or chemical phenomenon. In his first turn, Juho initiates the discussion byproposing a cause “ I think it’s like this…originally there was sugar or its carbonsystem, and this water separately and now they are bound to each other” . Juho’sdescriptive explanation indicates that he is approaching dissolving as a process. Inturn two, Jarmo continues by proposing a result “so there was a chemical reaction”(turn 2). Jarmo’s causal explanation reveals that his interpretation of dissolving refersto a chemical phenomenon. In the next turn, Juho extends the explanation by continu-ing “dissolved from them there has come third substance this is not pure sugar butis sugar-water” (turn 3). When explicating his point of view by using evidence, Juhocreates a situational meaning “sugar-water” while referring to the dissolving sample.In the next interaction, the group tries to resolve whether dissolving is a chemicalphenomenon by modelling.


4.1.2. Episode 3Here the students further their discussion about whether dissolving is a physical

or chemical phenomenon. This episode highlights the emergence of differentapproaches and explanations in the construction of a joint meaning. The episode isinitiated by the teacher saying, “which phase are you in right now” (turn 19). Thisinitiation is continued by Elina’s evaluative turn “we are just reasoning” (turn 20).Juho in turn 21 continues by proposing a result for dissolving “probably there is athird substance” . The teacher tacitly remains silent, realising that a response mightclose the reasoning process. This strategy appears to lead Jarmo to summarise thegroup’s joint interpretation of the situation “salt and water have dissolved..there isno salt and water but saltwater” (turn 22). Elina, on the other hand, in her extendingturn 23 contradicts the group’s solution by making a causal explanation “yes but wecould say that it is a solution of sodium chloride in water…there doesn’ t necessarilyexist a third substance it is still a solution” . In her turn 24 Paula extends the perspec-tive brought up by Elina by bringing in the interaction between metal and acid. Juhoagrees with Paula’s explanation in turn 25 and the students shift to discussing theirworking strategies, with the actual problem put aside for a moment.

4.1.3. Episode 6In Episode 6 the students finally construct a joint meaning for dissolving. This

episode highlights nicely the collaborative nature of interaction in which turns arewoven closely together. The thematic episode is initiated by Juho as he evaluateshis understanding of the situation, “ I don’ t know what dissolving is, whether onesubstance gives something to another” (turn 33). In turn 34 Paula continues evaluat-ing her learning experiences, “you can never read about it” . Jarmo joins in the dis-cussion by bringing in his learning experiences which appear to contradict Paula’s,“many times in high school but I have forgotten” (turn 35). Elina continues thediscussion further by proposing a cause for dissolving, “…on the other hand in dis-solving it seems to me that the crystal structure gets smaller in salt so that thesodium” (turn 36). Juho continues Elina’s elaboration by saying, “sodium chloridebreaks down” (turn 37). Elina extends the explanation building in turn 38, “sodiumand chloride those ions go there” in which she also brings the concept ion into theproblem-solving process. In his turns 39 and 41 Jarmo summarises the explanationfor dissolving, “yes the idea is like that those crystals in a way break down” (turn39) and “and when water is being evaporated those ions join together again” (turn41). In Elina’s turns 40 and 42 she supports Jarmo’s reasoning by agreeing with hisexplanations, “yes” (turn 40) and “ that’s how it seems to be” (turn 42). However,in this interaction cycle Elina’s turn 36 seemed to give rise to the final process ofexplanation building. In his last turn 43 Jarmo supports their joint explanation processby validating and adjusting the groups’ explanation within the framework of chemis-try.

4.1.4. The level and nature of participation within the groupThis section further highlights the level and nature of the students’ participation

in social explanation-building. The analyses demonstrate the presence of different


approaches and perspectives which were brought into the discourse by the membersof the student group.

Table 5 shows the distribution of discourse moves among the students in theirsocial activity during which the students elaborated the meaning for dissolving. Thedata demonstrate that the students’ discourse within the group was coherent andcollaborative in nature, often characterised by initiation, continuation, and extendingmoves. The data also indicate a rather symmetrical participation structure betweenthe students. However, one group member, Paula, seems to have been more silentthan others. Yet, her contribution was observed to be collaborative and extendingin nature.

Table 6 highlights the nature of the students’ logical processes in their discourse.The data analysis reveals that the group discourse was evaluative in nature (32.3%).Causes (25.8%) and results (22.6%) were also proposed in the students’ collaborativeinquiry. The data also show some differences between the students. Juho was foundto engage in proposing causes, whereas Elina seemed to evaluate joint problem-solving slightly more than others. Paula’s contribution to the group discourse wasdemonstrated in that she advanced evidence and evaluated joint activity.

Table 7 indicates that the students’ discourse was highly explicatory in nature.Forty-six percent of the students’ conversational turns were identified as beingexplicatory. Particularly causal explanations (40%) and formal explanations (35%)seemed to be the most frequent ones. Descriptive and everyday explanations werealso identified in the students’ social interaction (see Juho and Jarmo). The data alsoshow that three of the students, Juho, Jarmo and Elina, were all eager to explaindissolving. Paula, however, did not participate much in explanation-giving.

Table 8 shows that the students used diverse cognitive strategies in their problem-posing and solving. These strategies were posing a problem (constructing a question,raising a new question), using evidence, and applying a principle to a case. The dataalso indicate that there were student-specific strategies which appeared to supportthe students’ explanation-building. For example, Juho was eager to use evidence andconstructing questions. Jarmo, on the other hand, had a tendency to use everydayknowledge in his reasoning, whereas Elina applied principles to cases several times.The participatory structures of Paula’s cognitive strategies were using evidence andapplying a principle to a case. Although the frequency of Paula’s strategies was ratherlow in social interaction, the nature of these strategies implies that her approach wascritical in nature.

4.2. The nature of the students’ explanations in the pre- and post-test conditions

The results concerning the nature of the students’ conceptual explanations fordissolving in the pre-test and post-test conditions are discussed here under two sec-tions. First, the results from the case group will be interpreted. This is followed bya discussion concerning the data gathered from the whole science class.


Tab

le5

Dis

cour

sem

oves

(N=4

2)an

dth

eir

dist

ribu

tion

amon

gth

egr

oup

mem

bers

Nam

eD

isco

urse

mov

es

Initi

atin

gC

ontin

uing

Ext

endi

ngR

efer

ring

back

Agr

eein

gR

eply

ing

Com

men

ting

Con

clud

ing

Tot

al

Juho

34

22

213

(31.

0%)

Jarm

o1

51

11

21

12(2

8.6%

)E

lina

12

72

113

(31.

0%)

Paul

a2

24

(9.5

%)

Tot

al5

(11.

9%)

13(3

1.0%

)12

(28.

6%)

3(7

.1%

)5

(11.

9%)

1(2

.4%

)2

(4.8

%)

1(2

.4%

)42

(100

%)


Table 6The nature of logical processes (N=31) in the students’ discourse

Name Logical processes

Proposes a Proposes a Advances Suggests a Evaluates Contradicts Totalcause result evidence method

Juho 5 1 1 1 1 9 (29.0%)Jarmo 1 3 1 3 8 (25.8%)Elina 2 3 5 1 11 (35.5%)Paula 2 1 3 (9.7%)Total 8 (25.8%) 7 (22.6%) 3 (9.7%) 2 (6.5%) 10 (32.3%) 1 (3.2%) 31 (100%)

Table 7The nature of explanations (N=20) and their distribution among group members

Name Nature of explanation

Descriptive Causal Everyday Formal Totalexplanation explanation explanation explanation

Juho 1 3 1 1 6 (30%)Jarmo 1 3 2 6 (30%)Elina 3 3 6 (30%)Paula 1 1 2 (10%)Total 1 (5.0%) 8 (40%) 4 (20%) 7 (35%) 20 (100.0%)

Table 8Cognitive strategies (N=25) and their distribution among the group members

Name Cognitive strategy

Constructing Raising a Using Applying a Using Totala question new evidence principle to everyday

question a case knowledge

Juho 3 5 8 (32%)Jarmo 2 2 3 7 (28%)Paula 2 1 3 (12%)Elina 1 2 4 7 (28%)Total 4 (16%) 2 (8%) 11 (44%) 5 (20%) 3 (12%) 25 (100.0%)

4.2.1. Case groupTable 9 highlights the nature of the students’ explanations for dissolving in the

case group before and after working in the social science-learning situation.The data in Table 9 shows interesting connections between the students’ expla-


Table 9The nature of the students’ explanations in pre-test and post-test conditions

Student Explanations for dissolving in a Explanations for dissolving in apre-test condition post-test condition

Paula The substance dissolves into Physical phenomenon, where thesomething, e.g. salt into water, substance is dissociated intoso that there will be equilibrium different ions, e.g.in the liquid NaCl→Na++Cl-

Definition Descriptive explanation Formal explanationElina Real-life conditions remind me Dissolving happens when some

of the concepts watersoluble and solid substance is dissociated intoliposoluble and possibilities to liquid as structural parts, e.g.mix substances, mixtures, into moleculessolutions

Definition Practical explanation Formal explanationJarmo Solubility helps our life to a Substances differ in their

great extent, sugar in the coffee solubility, sugar is soluble incup, tablet in the mouth water, whereas sand remains

separate from waterDefinition Practical explanation Practical explanationJuho Dissolving means that two or Two or more substances are

more substances fuse at an dissolved into a new substance,atomic level so that a new there will be new moleculessubstance is formed

Definition (misconception) Explicatory explanation Explicatory explanation(misconception)

nations in the social learning situation and in solo situations. Jarmo’s explanationswere often practical in nature when he was participating in group discourse, and hiscognitive strategies rested on everyday knowledge. Although Jarmo expressed formalexplanations in the social learning situation, this type of explaining did not appearto emerge in solo situations. Juho, on the other hand, was eager to give descriptiveand causal explanations in group discourse, which was also reflected in his soloactivity. Although the group constructed a joint meaning for dissolving as a physicalphenomenon, Jarmo based his explanation on everyday experiences. Both Paula andElina were observed to have processed their explanations through their social interac-tions in the group. Whereas Paula defined dissolving in descriptive terms in the pre-test condition, her explanation in the social learning situation and in the post-testcondition started to reflect formal conceptualisation of the situation. In the pre-testcondition Elina based her explanation on everyday experiences. The social learningsituation and post-test condition suggest that she shifted her perspective in favourof a scientific one.

4.2.2. Whole science classThe analysis of the whole data shows that the students’ explanations for dissolving

were mainly descriptive and practical in nature in the pre-test condition, suggesting


that the students approached the topic on the basis of their everyday understandingof the phenomenon. This finding could be explained by the fact that although solu-bility was a familiar topic for all the students, they had not recently been involvedin elaborating and investigating the topic in an institutionalised learning context.

The post-test results suggest that the social science-learning situation gave thestudents opportunities to elaborate their conceptions of dissolving. The data showsthat the students’ explanations were also explanatory in nature in the post-test con-dition, including explanations which proposed results, causes and scientific defi-nitions. Thus, the social learning situation appeared to provide students’ with opport-unities to elaborate and extend their explanations. In addition, the data highlight thediversity in the students’ explanations, reflecting the multi-dimensional nature of thelearning situation which gave rise to the construction of different interpretations evenfrom the scientific point of view. Table 10 shows the results of the nature of thestudents’ explanations comparatively between the pre- and post-tests.

5. Discussion

Although explaining plays a crucial role in typical science-learning situations, verylittle is known on the processes involved in explanation-building in learning situ-ations wherein students are given opportunities to investigate scientific phenomenafrom different perspectives. In this study student-initiated explanations of dissolvingwere investigated in an instructional setting which was based on collaborativeinquiry. In the learning situation designed for the present study the grounding ofexplanation-building rested on informal or formal conceptualisation of the learningsituation. In this context, science learning could be participated in from differentframeworks with diverse interpretations (Kelly & Green, 1998). A significant con-dition for collaborative inquiry was that the diverse explanations emerging in theevolving interactions were adjusted to create a joint understanding of the topic inquestion.

The results derived from this study suggest that discourse processes in whichdifferent perspectives, interpretations and attitudes are negotiated create the groundsfor collaborative inquiry (Teasley & Roschelle, 1993). In this study, features of a

Table 10The nature of the students’ explanations for dissolving in the pre-test and post-test conditions

The nature of explanation Pre-test condition Post-test condition

Descriptive 9 4Practical 8 1Explicatory� Proposes a result 4� Proposes a cause 1 6� Formal explanation 3Total 18 18


shared conceptual structure were manifested in the logical processes of interaction,in the nature of explanations, and in cognitive strategies which were negotiated insocial activity. The diverse perspectives emerging in social interaction seemed to besupported by the fact that the group members differed in their approaches and per-spectives towards collaborative inquiry. For example, in the case study group Paulawas a critical participant who could be characterised as a silent thinker who onlytook part in the discourse at critical points to advance the group’s problem solving.Juho, on the other hand, approached the collaborative inquiry by “ thinking aloud”while making his thinking transparent to the group members. Jarmo seemed to furtherthe group’s thinking with a practical approach by adjusting informal and formalunderstanding for joint problem-solving. Elina approached collaborative inquiry fromthe perspective of formal understanding in extending and evaluating the groups’problem-solving processes. On these findings we can base the observation that het-erogeneity in the group seemed to create ideal conditions for collaborative problem-solving (Hogan et al., 2000; Howe, Tolmie, Anderson, & MacKenzie, 1992).

The study highlights the importance of value and respect for each other’sapproaches and contributions while sustaining collaborative inquiry. The presenceof these elements were identified in this study in the nature of the students’ communi-cative processes. For example, the analysis of discourse moves reveals rather har-monic turn taking patterns, starting from an initiation move which was often followedby continuing or extending moves. The triadic discourse episode, i.e. initiation–reply–evaluation, typical of teacher-directed classroom interaction (Sinclair & Coul-thard, 1975) was found to be more rare in the student-initiated group discourse, inwhich new thematic episodes emerged in the ongoing discourse where the studentsbrought in new perspectives and started to negotiate them.

The study sheds light on the reciprocal relationship between the communicativeand cognitive processes of discourse in collaborative inquiry. From the viewpointof explanation-building the case study shows that extending moves, particularly thosewhich proposed causes and results, advanced evidence and evaluated collective prob-lem-solving, seemed to be related to formal and causal explaining. The cognitivestrategies, often applied in the construction of formal and causal explaining, wereusing evidence and applying a principle to a case. These strategies seemed to beconnected with the students’ logical processes, namely proposing causes and results.The construction of everyday explanations was found to be connected with usingeveryday knowledge. Although these findings need further investigation owing tosmall sample size, the synchronic analysis, as demonstrated here, is likely to bringnew insights into the complexities of students’ reasoning processes in collabor-ative inquiry.

The results of this study indicate that the science-learning situation designed forthis study supported the students’ conceptual elaboration of dissolving. The analysesof the students’ pre- and post-tests suggest that there were qualitative changes in thenature of the students’ explanations. Whereas in the pre-test condition the students’explanations for dissolving were mostly descriptive in nature, in the post-test con-dition the explanations were more explanatory, reflecting cause–effect reasoning andformal reasoning. On the basis of these findings, it could be suggested that collabor-


ative inquiry gave rise to the construction of alternative explanations around theconcept of solubility.

The results of this study suggest that personal theories and explanations are notnecessarily replaced by formal explanations in collaborative interactions. Yet, collab-orative interactions and different perspectives which are negotiated in the flow ofsocial activity can give rise to the elaboration of personal explanations. In this elabor-ation process, everyday explanations and scientific explanations do not contradicteach other. Instead, they can be seen as complementary, offering students tools andperspectives to participate in different culturally based situations. A greater challengeis then to learn how and in what way to participate in different social contexts.

6. Conclusions

The study demonstrates the value of process analyses in investigating and evaluat-ing the nature of explanation-building processes in collaborative inquiry. The pro-cesses of negotiation of diverse perspectives and explanations were unravelled withthe help of an analytic method developed specifically for the present study. Themethod takes a synchronic and a diachronic approach to the analyses of collaborativeinteraction: on the one hand, the analysis method investigates group discourse acrossanalytical frames, while on the other hand, the method investigates the evolution ofgroup discourse on a moment-by-moment basis. The strengths of a multi-facetedmicro-level analysis of collaborative interaction appear to be in parallel with theexamination of discourse data. The different frames of the analysis support oneanother. Whereas the analysis of communicative processes reveals the beginningsand closures of interaction cycles, the analysis of explanatory processes gives mean-ing and context to them. The next challenge for the analysis framework is to developit to be suitable for different science-learning contexts across age-levels.

In summary, the present study raises important future research questions. Theoreti-cally, from the viewpoint of conceptual development, it seems important to investi-gate students’ conceptualisation processes in differentiating and integrating alterna-tive explanations into a common theoretical framework while learning chemistryacross different learning themes. Pedagogically, there is a need for further instruc-tional design to develop science-learning contexts which are realised in collaborative,student-initiated and inquiry-based working modes. Further, there is a need todevelop appropriate evaluation tools to understand the practice of collaborativeinquiry in the learning of science. The analysis method described and used in thisstudy could be viewed as an attempt towards this goal.

Acknowledgements

The study reported in this paper was supported by the Academy of Finland (Projectno: 41969). The authors would also like to express their gratitude to the anonymousreviewers for their valuable comments on the earlier version of the manuscript.


References

Barnes, D., & Todd, F. (1995). Communication and learning revisited, making meaning through talk.Portsmouth, NH: Boynton/Cook Publishers/Heinemann.

Bruner, J. (1996). The culture of education. Cambridge: Harvard University Press.Carey, S. (1985). Conceptual change in childhood. Cambridge MA: MIT Press.Chi, M. T. H., Slotta, J. D., & De Leeuw, N. (1994). From things to processes: a theory of conceptual

change for learning science concept. Learning and Instruction, 4, 27–45.Cobb, P., & Bowers, J. (1999). Cognitive and situated learning perspectives in theory and practice. Edu-

cational Researcher, 28, 4–15.Cole, M. (1996). Culture in mind. Cambridge, MA: Harvard University Press.Coleman, E. B. (1998). Using explanatory knowledge during collaborative problem solving in science.

The Journal of the Learning Sciences, 7 (3–4), 387–427.Crawford, T., Kelly, G. J., & Brown, C. (2000). Ways of knowing beyond facts and laws of science: An

ethnographic investigation of student engagement in scientific practices. Journal of Research inScience Teaching, 37, 237–258.

Driver, R., & Easley, J. (1978). Pupils and paradigms: a review of literature related to concept develop-ment in adolescent science studies. Studies in Science Education, 5, 61–84.

Gee, J. P., & Green, J. L. (1998). Discourse analysis, learning, and social practice: a methodologicalstudy. Review of Research in Education, 23, 119–169.

Herrenkohl, L. S., & Guerra, M. R. (1998). Participant structures, scientific discourse, and student engage-ment in fourth grade. Cognition and Instruction, 16 (4), 431–473.

Hogan, K., Nastasi, B., & Pressley, M. (2000). Discourse patterns and collaborative scientific reasoningin peer and teacher-guided discussions. Cognition and Instruction, 17, 379–432.

Howe, C., Tolmie, A., Anderson, A., & MacKenzie, M. (1992). Conceptual knowledge in physics: therole of group interaction in computer supported teaching. Learning and Instruction, 2, 161–183.

Howe, C. J., Tolmie, A., Greer, K., & Mackenzie, M. (1995). Peer collaboration and conceptual growthin physics: task influences on children’s understanding of heating and cooling. Cognition and Instruc-tion, 13, 483–503.

Kelly, G. J., & Chen, C. (1999). The sound of music: constructing science as sociocultural practicesthrough oral and written discourse. Journal of Research in Science Teaching, 36, 883–915.

Kelly, G. J., & Green, J. (1998). The social nature of knowing: toward a sociocultural perspective onconceptual change and knowledge construction. In B. Guzzetti, & C. Hynd, Perspectives on conceptualchange: multiple ways to understand knowing and learning in complex world (pp. 145–181). Mahwah,NJ: Erlbaum.

Kumpulainen, K. (1996). The nature of peer interaction in the social context created by the use of wordprocessors. Learning and Instruction, 6, 243–261.

Kumpulainen, K., & Mutanen, M. (1999). The situated dynamics of peer group interaction: an introductionto an analytic framework. Learning and Instruction, 9, 449–473.

Lemke, J. L. (1990). Talking science: language, learning and values. Norwood, NJ: Ablex.Mead, G. H. (1934). Mind, self and society. Chicago: Chicago University Press.Rogoff, B. (1990). Apprenticeship in thinking: cognitive development in social context. New York: Oxford

University Press.Roth, W.-M., & Lucas, K. B. (1997). From “ truth” to “ invented reality” : a discourse analysis of high

school physics students’ talk about scientific knowledge. Journal of Research in Science Teaching,34, 145–179.

Russell, T. (1983). Analyzing arguments in science discourse: can teachers’ questions distort scientificauthority? Journal of Research in Science Teaching, 20, 27–45.

Schauble, L., Glaser, R., Duschl, R. A., Schulze, S., & John, J. (1995). Students’ understanding of theobjectives and procedures of experimentation in the science classroom. The Journal of the LearningSciences, 4, 131–166.

Schnotz, W., Vosniadou, S., & Carretero, M. (1999). New perspectives on conceptual change. Amsterdam:Pergamon Press.


Sfard, A. (1998). On two metaphors for learning and the dangers of choosing just one. EducationalResearcher, 27, 4–13.

Sinclair, J. McH., & Coulthard, R. M. (1975). Towards an analysis of discourse. The English used byteachers and pupils. London: Oxford University Press.

Saljo, R. (1999). Concepts, cognition and discourse: from mental structures to discursive tools. In W.Schnotz, S. Vosniadou, & M. Carretero, New perspectives on conceptual change (pp. 81–90).Oxford: Elsevier.

Teasley, S., & Roschelle, J. (1993). Constructing a joint problem space: the computer as a tool for sharingknowledge. In S. P. Lajolie, & S. J. Derry, Computers as cognitive tools (pp. 229–257). Hillsdale,NJ: Lawrence Erlbaum.

Viennot, L. (1979). Spontaneous reasoning in elementary dynamics. European Journal of Science Edu-cation, 1, 205–221.

von Glasersfeld, E. (1989). Constructivism in education. In T. Husen, & T.N. Postlethwaite (Eds.), Theinternational encyclopedia of education (pp. 162–163). Oxford: Pergamon Press.

Vosniadou, S. (1994). Capturing and modelling the process of conceptual change. Learning and Instruc-tion, 4, 45–69.

Vosniadou, S. (1999). Conceptual change research: state of the art and future directions. In W. Schnotz,S. Vosniadou, & M. Carretero, New perspectives on conceptual change (pp. 3–13). Oxford: Elsevier.

Vygotsky, L. S. (1962). Thought and language (E. Hanfmann, & G. Vakar, Trans.). Cambridge, MA:MIT Press.

Vygotsky, L. S. (1978). In M. Cole, V. John-Steiner, & E. Souberman, Mind in society: the developmentof higher mental processes. Cambridge, MA: Harvard University Press.

Watts, D. (1991). The science of problem solving. A practical guide for science teachers. Portsmouth: Hei-nemann.

Wells, G. (1999). Dialogic inquiry. Towards a sociocultural practice and theory of education. Cambridge:Cambridge University Press.

Wenger, E. (1998). Communities of practice. Learning, meaning, and identity. Cambridge: CambridgeUniversity Press.

Wertsch, J. (1991). Voices of the mind: a sociocultural approach to mediated action. Cambridge, MA:Harvard University Press.

Collaborative inquiry and the construction of explanations in the learning of science

Documents