Individual effort in field and laboratory experiments is sensitive to information about how others perform. Typically one group reference point is provided in these experiments and so there is only aspect to social comparison. However, in many natural settings, indeed sometimes as a result of public policy intervention, there are several such group reference points and this potentially complicates the kinds of social comparison that can be made. This paper reports on an experiment where there are such multiple group reference points and it examines how individual behavior is affected when there is this richer scope for social comparison. We test whether multiple reference groups affect aggregate performance and its variance within and between each group and whether any of these effects depend on the mechanism generating group affiliation.
|Publication status||Unpublished - 2014|