Finally, we found evidence for solidarity imbalances. While non-Black participants expected Black acquaintances to be especially supportive, Black participants expected lower support from non-Black acquaintances for both racism and sexism.

We also found that past solidarity behavior increased expectations for support. Past solidarity behavior was an especially important cue if the person belonged to an outgroup and when the behavior was especially active (e.g., protesting).

Overall, expected support followed these patterns: Black > Hispanic > Asian ≥ White & women > men. Racial/ethnic ingroup status mattered more for racism while a person’s gender mattered more for sexism.

In 2 experiments (N = 2138), participants learned about acquaintances who varied in race/ethnicity, gender, & past solidarity. We asked whether those acquaintances would become angry on their behalf & take action to support them after someone else said something racist/sexist.