What motivates people to discriminate against outgroups? Why are some people more likely to discriminate than others? Several theories propose that low self-esteem predicts discrimination. However, empirical research in support of these theories has been decidedly mixed. The current research examines whether the association between self-esteem and discrimination can be clarified by examining implicit, in addition to explicit, measures of collective self-esteem (CSE). I hypothesized that the interaction between implicit and explicit CSE would predict intergroup bias such that people with inconsistent implicit and explicit CSE would exhibit more intergroup bias than people with consistent implicit and explicit CSE. I reasoned that people with inconsistent CSE would be especially motivated to discriminate as a way to reduce the discrepancy between their implicit and explicit attitudes.

Across two studies I examined whether implicit and explicit CSE interact to predict intergroup bias, as indexed by: attributions about, evaluations of, and resource allocation to the ingroup versus outgroup. Study 1 examined White and Latino participants' CSE regarding their university and ethnicity. Study 2 examined only White participants' CSE regarding their university, but introduced a manipulation of ingroup threat (article about university's poor reputation) and controlled for related variables (implicit and explicit personal self-esteem and explicit group identification).

The results of the current research did not support predictions. Overall, the findings were mixed; some of the results suggested that participants with consistent CSE exhibited more intergroup bias, whereas other results suggested that participants with inconsistent CSE exhibited more intergroup bias. In study 1, Latino and White participants with consistent CSE evaluated the ingroup, relative to the outgroup, more positively compared to participants with inconsistent CSE. In contrast, Latino and White participants with inconsistent CSE allocated more resources to the ingroup versus the outgroup. In study 2, in contrast to study 1, implicit and explicit CSE interacted to predict intergroup bias among White participants, but only when they experienced a threat to their university ingroup. The pattern of the interaction in the threat condition, however, paralleled some of the results of study 1. White participants with consistent CSE evaluated the ingroup, relative to the outgroup, most positively.

Overall, the current research suggests that the relationship between implicit CSE, explicit CSE, and intergroup bias is complex and may vary depending on a multitude of factors. Nonetheless, despite inconclusive findings, the current research demonstrates the importance of examining both implicit and explicit attitudes when exploring the processes underlying discrimination.