We examine whether being included in or excluded from a group based on a given dimension of one's identity increases the salience of this identity dimension in unrelated interactions. Using a laboratory and a hypothetical online experiment, we study two abstract and ex-ante equally relevant dimensions of identity as well as two relevant identity dimensions.

Participants are exposed to a situation where one member of a three-member team is excluded, ostensibly because they do not share a given dimension of their identity with the other two (included) members of the team. We subsequently measure the ingroup bias of all participants on both identity dimensions.

While we find in the laboratory experiment for the abstract identities that there is no effect of exclusion, being included increases ingroup bias on the identity dimension that was the basis for inclusion. To check for the robustness of these results, we conduct an additional online experiment in which we compare the effect regarding the abstract identities with real and relevant identities.