Warmth and competence in implicit stereotypes and discrimination

Abstract: It is well established that we do not judge other people on a one-dimensional scale (i.e., good - bad), but rather based on two fundamental dimensions. The first dimension is warmth, which essentially answers the questions of what the other person’s intentions are (e.g., friendly or malicious). The second dimension is competence, which in contrast answers the question regarding the person’s capability to carry out those intentions. The stereotype content model (SCM; Fiske, Cuddy, Glick & Xu, 2002) suggests that these two dimensions are not only relevant for person-perception, but also for intergroup bias. Some groups are stereotyped as both warm and competent (e.g., the majority population or the ingroup) or neither warm or competent (e.g., an ethnic minority). Some groups have mixed stereotypes: warm, but not competent (e.g., Greeks), or competent but not warm (e.g. Germans). There is already considerable research supporting the SCM. Yet, these studies have so far focused almost exclusively on self-reported stereotypes and prejudice. The present thesis aimed to extend this research by focusing on two unexplored areas where a warmth and competence perspective might prove useful: implicit stereotypes and discrimination. Study I showed that it is possible to capture mixed stereotypes (in terms of warmth and competence) using the Implicit Association Test (Greenwald, McGhee & Schwartz, 1998). The study also provides some preliminary indication of the usefulness of this approach. Specifically, whereas the implicit measures were sensitive to ingroup bias, the explicit measures were not. Study II demonstrated that mixed stereotypes translate into mixed discrimination. Across two experiments, groups that are stereotyped as warm but not competent (preschool teachers and Greeks) were discriminated in a competence paradigm, but favored in an empathy (warmth) paradigm. In contrast, groups who are stereotyped as cold but competent (lawyers, Germans) were favored in the competence paradigm, but discriminated in the empathy paradigm. Importantly, a one-dimensional perspective failed to find any indication of discrimination, since the groups were treated equally if the two dimensions were collapsed. Study III investigated real-life hiring discrimination in a field experiment. Fictive applications were sent to 5,636 job openings. By experimentally varying whether the applicant had an Arab or Swedish sounding male name, and whether he appeared warm and/or competent in the personal letter, we were able to investigate how individuating information related to warmth and competence interacts with ethnic hiring discrimination. We found substantial discrimination in that Arab applicants received fewer invitations to job interviews. Consistent with the stereotype content of Arabs, an applicant with an Arab sounding name had to appear both warm and competent in order to increase his chances. Interestingly, in order to be on (almost) equal terms as an applicant with a Swedish sounding name, he had to be both warmer and more competent. In conclusion, the present thesis suggest that much is to be gained from viewing intergroup bias from a two-dimensional perspective, regardless if the focus is on implicit stereotypes or on discrimination. Researchers in implicit stereotypes who do not consider warmth and competence may haphazardly conclude that a group is either negatively or positively stereotyped, when the implicit stereotype is actually mixed (e.g., warm but not competent). The same is true for discrimination, which can also be mixed in terms of warmth and competence. Furthermore, even when studying discrimination toward groups that are clearly stereotyped as altogether bad, a one-dimensional perspective may still lack precision. Indeed, it was not enough for an applicant with an Arab sounding name to appear warm or competent: he had to appear simultaneously warm and competent if he were to increase his chances to receive job interviews. Hence, it would appear that intergroup bias is a too complex phenomenon to be understood as simply good or bad.