Frans Ronteltap

Maastricht Learning Lab

University of Maastricht

The Netherlands

F.Ronteltap@mmi.unimaas.nl

Collaborative learning attracts a lot of attention at this moment. Meta-studies of effects of cooperation and collaboration during learning, in which small-group learning was compared with individual learning, demonstrated better results in terms of achievement, motivation and successful completion of courses. (Springer, Stanne & Donovan, 1999). The technological revolution is the second driving force in the growing interest for computer supported collaborative learning. However, in this context it is remarkable that to a much smaller extent the contribution of collaborative work has been a topic of research. The following quote from Gerry’s working paper was for me the immediate cause to join this workshop: “Despite frequent references to constructivism in the CSCL literature, it is not clear which cognitive processes are involved in collaborative knowledge building”. But we need to know this in the design and development of tools that will support the learner(s).

A detailed analysis of the interactions in collaborative learning environments might give the answer to the following basic questions:

1.     To what extent can changes in the representation of knowledge representations be seen as a result of collaboration in the learning process?

2.     What type of interactions in collaborative learning are responsible for these changes in knowledge representation?

The evaluation of an experiment where students worked in a learning environment (POLARIS) that was developed in Lotus Notes triggered off those two questions for me. The students who participated in these experiments were used to participate in tutorial groups in a problem based curriculum. At the end of a meeting students define a set of learning issues to study in the time period before the next meeting. All issues are studied by all students, but they are encouraged to use different sources for information. In the POLARIS environment they could write a document, and their peers could read and comment those documents. The interface of the database views was the traditional threaded discussion view. As early as in the design phase of this learning environment, we anticipated the need for the development of a set of survey tools that students could use for navigation, to control the amount of incoming information and to derive benefit from the possiblity of access to their documents. With the Questions & Answers button they could produce a new document that contained all questions and answers. With the References button all references of the documents could be studied. The Report button presents the contributions of all students concerning one type of report or keyword. And the Thread button combines the content of all documents in a thread. In all cases students could copy the results of these actions to their individual directories.

In interviews students told us, however, that these tools were seldom used. This finding fits in the regular way of group learning to which they are used. In peer meetings they present and discuss their results. And after these presentations they start with a new case. The function of the group meeting for students was regularly supporting their individual learning process, and they continued to do so in the new environment. The basic cognitive process is elaboration or articulation (Chi, Bassok, Lewis, Reimann & Glaser, 1989). We gave them tools that could (possibly) support the construction of shared knowledge, but they did not pick it up. The first thing that we learned from this finding was that, if we require students to collaborate in the way of knowledge building, we need to direct them in that way: transcript based assignments, tutor interventions, but also technical adaptations. 

At this moment we revise the functional design of the environment and will adapt the knowledge building part of the environment. For example, we are considering the following adaptations:

·       Specification of reaction types. Close reading of the feedback documents in the first experimental databases learned us that students produce three basic types of reactions: comments, supplements and questions (and answers). This corresponds to what Wegerif, Mercer & Dawes (1999) called three kinds of talk or “social modes of thinking”: disputational talk, cumulative talk and exploratory talk. In the next version of the KBE we will ask students to categorize their reactions, and develop a tool that links these different reactions by the user.

·       Encouragement of convergence. In the first version we did not gave students the possibility to express their agreement with the content when they read the work of peers. When they did not, they could comment of course. In the next version we will introduce a tool of agreement (adding names to documents of peers, or collaborative writing) and subsequently a tool to select a view that reflects the agreement in the KBE.

The last two examples illustrate the phase in which we are in the development of our learning environment at this moment.  I hope to get some inspiration and ideas in joining this group and would like to express my interest in the theme “Defining KBEs”, in particular the question “How do KBEs differ from discussion forums?”

·       Chi, M.T.H., Bassok, M., Lewis, R., Reimann, P., & Glaser, R. (1989). Self-explanation: How students study and use examples in learning to solve problems. Cognitive Science, 13, 145-182.

·       Springer, L., Stanne, M.E., & Donovan, S.S. (1999). Effects of small-group learning on undergraduates in science, mathematics, engineering, and technology: A meta-analysis. Review of Educational Research, 69 (1), 21-51.