Back to top

Group Brainstorming in Organizations: Implementing the Functional Theory of Group Decision-Making as a Means for Increasing Performance

Heuett, Kyle B
Committee Members: 
Dr. Kenneth Levine
May 2015


Brainstorming was first introduced as a group focused method for generating ideas on behalf of an organization. Past studies on brainstorming have been inconclusive about the effect of certain types of brainstorming techniques on the number of ideas and the quality of ideas generated by groups. In seeking to develop different techniques for brainstorming research has lacked a theoretical guide that has led to mixed results at best about different brainstorming techniques. Further, brainstorming research conducting using experimental methods have lacked realism compared to industrial groups; specifically this lack of realism is evident in the history of brainstorming groups and the topic given to brainstorming groups. This study introduced the functional theory of group decision-making as a means of addressing issues of theory and realism and improving what is known about brainstorming performance. The functional theory allows groups to brainstorm according to five task requirements, the performance of these brainstorming groups can be compared against brainstorming groups using past techniques to determine the effect of different brainstorming techniques. Also, an extensive induction of group history was used for half the brainstorming groups prior to the brainstorming session. By doing this issues of realism can also be addressed. To further address realism in brainstorming groups a salient topic was selected for all groups to generate ideas about. Results indicate that history had a significant main effect on the number of ideas generated. Further, there were significant differences in the number of ideas generated across the different brainstorming techniques. Results were inconclusive on any differences regarding technique or history in regards to idea quality. However, a significant main effect was present for one technique across history and zero-history groups. Further results and theoretical implications follow.