Group size, member ability, and social decision schemes on an intellective task.

Abstract

510 college students first took Part 1 of the difficult Terman Concept Mastery Test as individuals and then retook the same test as control individuals or in cooperative high-ability or low-ability groups containing 2-5 Ss. 9 a priori social decision scheme models were tested as theories of the underlying group process. An a posteriori social decision scheme was also induced from a comparative previous study with 240 4-person groups and then tested against the current groups. The best-fitting a priori social decision scheme suggested an underlying "truth-supported wins" process, a finding which was more pronounced with increasing group size and for high-ability groups. However, the a posteriori principle, representing a combination of a truth-supported wins process, strong conformity pressures against a single correct member, and an increment from grouping when no member was correct, fit the current data better than any of the a priori social decision schemes.