Saturday, December 1, 2007

Computer-Supported Collaborative Learning

Technology and sociality are interrelated. Few social groups can function without some kind of technology to connect them, and technologies with no social function serve little purpose. Stevens discusses Computer-Supported Collaborative Learning (CSCL) and the relationships among people, tasks, and technologies in performing tasks. Roschelle, in discussing Convergent Conceptual Change (or collaborative learning), describes a process by which students can agree on a given piece of knowledge.

Roschelle’s four-featured process includes production of a deep-featured situation, interplay of physical metaphors, interactive cycles of conversational turn-taking, and progressively higher standards for convergence. In other words, they need the following: a case that captures their attention and requires some effort to accomplish a goal, a connection to their existing knowledge, iterations of practice and feedback, and assessment as they apply their newly constructed knowledge. So we’re talking about Gagné’s Nine Events of Instruction, with a few of the steps combined and applied to a technology-based collaborative environment.

I was interested by Steven’s observation that in a middle school classroom the students mirrored the practice in architectural firms of designers using paper and draftsmen translating the paper into a computer model, even though the computer was intended to be the primary design tool and the students had little training in paper-based designing. Once the roles were established, it was interesting that the students declined to rotate into the other roles as they had been told to do by their teacher. After reading this, I’ll have to take a hard look at my own role as R programmer in the group for my statistics class, where the other members write on the white board and discuss theories, while I actually enter the code based on their recommendations. Apparently I’m the junior draftsman of the group, but seriously with the number of group projects I’ve been in, it appears obvious that in almost any situation with teams, the roles seem to be established quickly and are often difficult to change later.

Steven’s discussion of assessment was very important, given the weight we place on assessment in courses. Students quickly figure out what is and isn’t important in a class, based on how they are assessed. The classroom designers learned very different skills, so how are they to be appropriately assessed? A test on using the CAD software would (and did) result in low scores for the designers, because scores were based on traditional uniform-experience assessment. In the reading I have done on Problem-based Learning (PBL), which this experience appears to match pretty closely, assessment of individuals is always a difficult issue. A possible method for grading is to rotate groups throughout the class, so each individual receives the average of their group scores, if there is time to do so. Assessment should be based on performance on the task, not on a True/False test afterwards. Although students involved in PBL often score lower on tests than those involved in didactic learning, they tend to retain what they did learn longer.

Steven’s discussion of division of labor and Roschelle’s discussion of collaborative learning do not compete, but are complementary. The roles that are played by group members depend on the type of environment in which the group is placed. Individual members of the group come to a shared understanding (hopefully), but the path to get there is dependent on the roles each member plays. An important part of PBL is allowing the group to come to consensus in whatever way works for them. They learn the subject matter, but they also learn processes for collaborating to make decisions and construct knowledge.

It is unfortunate that so much of traditional educational practice discourages teamwork, when the interaction while learning is what really encodes the concepts being studied. Collaboration is key. Some real-world environments continue to function on the competition model we are taught in school, but they are not as productive. That is not to say that competition is bad, because it’s not. It is good but in the right context. A goal, then, should be to provide technology that encourages collaboration, so that sociality can thrive.

Roschelle, J. M. (1992). Learning by Collaborating: Convergent Conceptual Change. Journal of the Learning Sciences, 2(3), 235-276.

Stevens, R. (2000). Divisions of labor in school and in the workplace: Comparing computer and paper-supported activities across settings. Journal of the Learning Sciences, 9(4), 373-401.

No comments: