Do Group Discussions Serve an Educational Purpose?

This study examines the effects of two information components on observers' ability to learn from group discussion. A social emotional and a factual information component were integrated in videotaped group discussions of a classical problem-solving task. To eliminate normative influence observers were employed to evaluate manipulated performance quality (good/bad) and group atmosphere (harmonious/conflictual). It could be demonstrated that without comparison level observers were unable to detect differences in performance quality but were secure in identifying the group atmosphere. With possibility for comparison of manipulated discussions the group atmosphere was used to evaluate performance quality revealing that learning was based on the wrong cues. Variation of both components showed that by increased complexity reference points were not present any more to correctly evaluate performance quality. Results suggest that wrong individual learning effects are considerably detrimental to group problem-solving quality