Abstract

The present study was designed to contrast two major distribution principles : the equity principle and the equality principle. It was proposed that group cohesiveness is an important condition which determines a person's preference of the distribution principles. It was also proposed that such preference would be effective both in allocation behavior and in reactions to injustice caused by imposing a non-preferred distribution priniple. The following two hypoheses were tested. (1) The equality principle will be preferred when group cohesiveness is high, but the equity principle will be preferred when group cohesiveness is low. (2) People will try to eliminate inequality rather than inequity when group cohesiveness is high, but they will try to eliminate inequity rather than inequality when group cohesiveness is low. In order to test the hypothesis(1) experiment I was conducted. In this experiment, subjects participated as a four-person unit and performed a simple task as a two-person group. Group cohesiveness was manipulated by assigning subjects to either group competition condition(high group cohesiveness condition) or individual competiton conditon(low group cohesiveness condition), and input level was manipulated by giving subjects false feedback on their task performance. The design was a group cohesiveness levels(high and low)×input level (high and low) factorial design and the dependent variable was the kind of the distribution principle employed when subjects allocated tokens to ingroup members. The results supported the hypothesis(1). The experiment II was conducted to test the hypothesis(2). In this experiment, subjects performed two tasks and were allocated outcomes for the first tasks according to either the equity principle or the equality principle, thus creating the state of injustice from the standpoint of one of the distribution principles. The subjects in this experiment were given opportunity to eliminate the state of injustice through adjusting their performance level(in put) for the second task. The dependent variable was the number of questions attemp ted in the second task. The results were consistent with the hypothesis(2). Thus, all the hypotheses were supported in the experiments, providing strong evidence for the relationship between group cohesiveness and the preference of distribution principles. The implications of the present study was viewed in the context of the equity principle theory versus multi-principle theory issue. Also, the limitations of the present study and the necessity of the further research were discussed.