Third International Conference on the Teaching of Psychology
Hokkai Gakuen University,
Sapporo, Japan


The purpose of this study was to examine the effects of group work training on group decision-making performance. The study was conducted in connection with a small class in group psychology at a university. The class was assigned certain group activities—such as cooperation for problem-solving—on purpose to improve student’s understanding of the psychological aspects of group processes. It was also expected that group activities would enable the students to develop their skills in group working. Students have few such opportunities during their regular university education in Japan. The study indicated the degree to which the students who engaged in group activities performed effectively on a decision-making task as a group.

Method: Participants were 135 undergraduates enrolled in a psychology class and 30 undergraduates registered for a class in group psychology at a university in Japan. Prior to the study, the latter 30 students had engaged in group activities for two semesters (trained condition). After each activity was finished, they were given a lecture on group psychology and asked to present a short paper. Students with no experience of such group activities numbered 135 (untrained condition). All the participants were randomly assigned to five-person groups in each condition and instructed that they would engage in a decision-making task. The number of groups in the untrained and trained conditions was 27 and 6, respectively.

The decision-making task was the NASA Moon Problem (Hall, 1971), wherein participants are asked to imagine that a spaceship has crash-landed on the moon and to rank order 15 items in terms of their relative importance for survival. Participants engaged in this task individually at first, and then made a decision as a group. On completion of the group decision-making task, participants had to estimate their familiarity with each group member.

Results and Discussion: The quality of a decision can be assessed in terms of its summed deviations from the expert rank order. This summed deviation represents an error score (0–112). When a group’s error score is smaller than both the average individual error score and the smallest individual error score in the group, the group is considered to have performed the task effectively.

In order to examine whether the trained groups performed more effectively than the untrained groups, we compared the frequency of groups in which the group’s decision was superior to that of its most proficient members in each condition. The results revealed that the frequency of effective groups in the trained condition (5 of 6) was higher than in the untrained condition (11 of 27). This might suggest that group work training had an impact on group performance.

A further analysis was conducted to investigate the degree to which familiarity with group members was related to group performance. As a result, it was observed that the frequency of effective groups in the familiar groups was higher than in the unfamiliar groups. These results imply that an increased closeness among group members through group work has an independent effect on group performance.

Back to index of presentations


Home Page

© 2008 Victor Karandashev