NOBODY REALLY KNOWS - THE STRUCTURE AND ANALYSIS OF SOCIAL CONSTRUCTIVIST WHOLE CLASS DISCUSSIONS

Increasing recognition of the social aspects of constructivist learnin g has led to calls for the greater use of discussion in science classr oom. Yet research on classroom discussion has shown that it is often n ot suitable for social constructivist aims. This paper presents a brie f outline of a method of conducting appropriate social constructivist discussions, plus a model for the structure of open-ended thinking thr ough such discussion, first advanced by David Perkins (1994), in which discussions are characterized as building a conception through the us e of epistemic games under the control of a structured heap. It then o utlines a method for the analysis and evaluation of these discussions, illustrated by excerpts from a classroom transcript. The methodology arose from a project to evaluate the processes of such discussion, and the impact on students' scientific reasoning, over a school year. Imp lications for researchers and curriculum planners are presented.