The use of achieved power in a probabilistic mixed-motive coalition situation

Abstract

This study demonstrated that achieved power could be used as a basis for creating individual differences among the participants in a probabilistic mixed-motive coalition formation situation. One hundred and thirty-five male subjects organized into forty-five triads participated in two competitive games. The first game was a practice game which gave the subjects the opportunity to form estimates of the capabilities of the members in their triad. Immediately following the practice game, pairs of subjects were permitted to meet in private in order to discuss whether they wanted to form a coalition. Two-person coalitions were allowed in the second game; and if a coalition had formed, the scores of the coalition partners were additively combined so that the coalition competed against the excluded third member of the triad. Either the individual winner of the second game or the winning coalition was given a chance in a raffle offering a $50.00 bonus. Points scored in the practice game as well as subjective appraisals obtained from participants in the form of ratings were used in comparable analyses to investigate the relationship between the initial distribution of power within each triad and coalition variables of interest. The general finding of the study was that a "strength is strength effect" was present with respect to coalition partner preferences. This finding supports the results of other studies which have found evidence of a "strength is strength effect" in probabilistic situations where assigned power was used. Therefore, the findings of this study using achieved rather than assigned power, substantiate the generality of the "strength is strength effect" in probabilistic situations. However, in this study several phenomena related to the "strength is strength effect" were discovered using subjective appraisals of power that would not have been evident had exclusive reliance been placed upon observable indications of power within the situation. Caution was advised against assuming that agreement exists among the participants' perceptions of power in probabilistic situations. The findings of the present study were discussed with reference to the inability of existing coalition theories to handle data obtained in probabilistic situations. Implications of the results and suggestions for future research were also considered.