Insufficient Justification for Exclusion Prompts Compensatory Behavior

ABSTRACT

This experiment examined participants’ compensatory behavior toward an excluded stranger. Participants engaged in a four-person social introductions task and rank-ordered each other member of the group; the lowest ranked person was excluded from a subsequent game. Using a 2 × 2 design, participants were randomly assigned to a justification condition (insufficient vs. sufficient) and to an exclusion responsibility condition (responsible for exclusion vs. exclusion by random selection). Results revealed that after limited introductions (i.e., insufficient justification for one’s ranking decision), being responsible for the exclusion prompted compensatory behavior toward the excluded stranger. However, after extended introductions (i.e., sufficient justification of one’s ranking decision), participants did not compensate the excluded person. These results suggest that insufficient justification for exclusion may lead to compensatory behavior, when one is responsible for the exclusion.

Keywords:

• compensatory behavior
• responsibility
• restitution
• sources of exclusion

Notes

1. Although highly similar, we acknowledge the difference between ostracism, rejection, and exclusion. Williams (2009) contends that ostracism involves both excluding and ignoring others, rejection involves a clear pronouncement that one is not wanted, and exclusion is the process of being separated or held apart from others. For the sake of clarity, and because our method most clearly aligns with exclusion, we will use the term exclusion for the remainder of the paper, though it is possible that some of the processes we describe apply to the other terms as well.

2. Inclusion or removal of the outlier did not substantively change the results; the key interaction was still significant if the outlier was left in the sample.

3. We pilot-tested this work allocation task on a separate sample of 59 undergraduate students. After imagining interacting with strangers for 5–10 minutes, participants allocated work based on rankings. As predicted, a repeated measures t test revealed that participants allocated themselves fewer division problems (M = 151.95, SD = 77.60) than those they ranked last (M = 200.42, SD = 93.75), t(58) = 2.51, p = .015, d = .56 (mean difference = 48.47, 95% CI [9.83, 87.12]).

4. We also analyzed the data by examining the number of problems assigned to the last-ranked team member as the primary dependent variable, and although there were no significant main effects, the same predicted interaction emerged as significant, F(1, 115) = 6.09, p = .015, partial η² = .050, as was the key simple main effect of responsibility in the insufficient justification condition, F(1, 115) = 7.37, p = .008, partial η² = .060 (mean difference = 26.44 problems, 95% CI[7.15, 45.74 problems]), demonstrating a similar pattern as the results we report.

Additional information

Notes on contributors

Daryl R. Van Tongeren

Daryl R. Van Tongeren, Lindsey M. Root Luna,

Charlotte VanOyen Witvliet

and Charlotte VanOyen Witvliet are affiliated with the Department of Psychology at Hope College.