Activity 4.2: Reading the Roschelle (1992) paper
Roschelle, J. (1992) 'Learning by collaborating: convergent conceptual change', Journal of the Learning Sciences, vol. 2, no. 3, pp. 235-76
We suggest that you divide up your reading of the paper in the following way:
- Read pp. 235-41 (up to the heading 'The Case Study').
Both 'collaborative learning' and 'conceptual change' are areas of research. As you might guess, 'collaborative learning' is about learning together but it is used by different authors in a variety of ways. (You may remember Hiltz and Meinke (1989) provided their own definition.) 'Conceptual change' refers to changes in people's understanding of concepts. There is a great deal of research into conceptual change in science learning. Roschelle gives his own definition of conceptual change on p. 237.
You might also want to note that 'CA' generally refers to Conversation Analysis. However, 'CA' is often referred to in the paper in the context of 'conversational actions' so that overall we cannot be clear as to Roschelle's exact meaning.
Theory - relational, situated view of learning
As you read consider the following questions:
- What argument is being made?
That students working collaboratively need to converge to develop a shared meaning and this process is characterised by four features:
Production of a deep features situation
Interplay of physical metaphors
Interactive cycles of conversational turn-taking
Progressively higher standards of evidence for convergence
- How does it relate to the kind of research being reported?
Students were working in pairs on a computer simulation to refine their concept of acceleration. The pair of students analysed were considered by the author to have achieved convergent conceptual change.
- How does the research question interact with the research work being undertaken?
Research question: How can the students converge on a deep new conception with only figurative, ambiguous and imprecise language and physical interactions at their disposal?
The study analysed the interactions between the students and between the students and the computer simulation.
- What kinds of evidence are relevant to the research question?
Tasks performed on the computer
Individual post-study interviews to determine whether the two students share the same conception.
- Read pp. 241-63 (up to the heading 'Evaluation of the Claims').
Read this section through fully before considering the following questions:
- How are the results/data presented? Compare with how Wegerif and Mercer presented their results in Reading 2.
Extracts from conversations shown in each paper however Roschelle emphasises the context and interactions with the computer and the student's body language and gestures which fits well with the research argument concerning the interactions.
- What does this say about the kinds of material that count as 'evidence' for the claims being made?
Conversation is not enough to analyse group interactions. He has added in body language, gesture and also physical metaphors such as Carol's demonstration on her hands.
- What does the author consider relevant about the context or setting for the study?
Private, urban high school
past knowledge: algebra
intellectual ability: average students but struggling with science
Relationship: close friends who had worked together on previous occasions
Timing: 2X one hour sessions after school
Analysis timing: after 12 problems where they had not yet developed an explanation that corresponded to a scientific understanding
- Read the remainder of the paper.
- What is the nature of the claims made in the evaluation?
A shared conceptual change occurred which was compatible with a scientific interpretation of velocity and acceleration.
Gradual convergence towards a shared understanding.
- How is the previously reported data used to support these claims?
Data from results section used to illustrate each part of the claims
- In what ways does the author relate the reported results to the wider literature?
Restates the basis of the problem with reference to literature
Majority of literature uses basis of contructivism but states that few theories account for convergent constructions in face of tendencies to diverge.
Tendency to diverge is strong in science (not backed by reference)
Discussion on how computer simulation can be truly constructive and situated or misused by teacher. How relational theory can assist this.
Differences from Vygotsky and Piaget and how this work supplements it
- Are the theoretical recommendations justified by the reported research?
'A case study cannot prove or disprove a theory, but it can clarify the meaning and import of a set of ideas'
'Mainstream conceptual change research on science learning should focus attention on convergence'
I believe that this is justified by the evidence produced here from one case study. The case study shows some good indications on how the two students collaborate to produce shared knowledge and it does need further work to see if this case is reproduced for other successful cases and also the difference with unsuccessful cases.