Suppressing stereotypes of the poor: rebound effects can be positive (as well as negative) : a thesis presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Psychology at Massey University, Auckland

Abstract

Suppressing unwanted stereotypes ironically leads suppressors to think and behave in a more stereotypical manner than controls. Suppression studies typically test for rebound with target groups that perceivers may feel entitled to stereotype (e.g., skinheads) or that are associated with negative stereotypes (e.g., African Americans). In contrast, stereotypes of the aid-related poor are expected to contain ambivalent content: a mix of both positive and negative stereotypic information. Since this content may affect perceivers’ cognitive processes of stereotype activation, application, suppression and rebound, it was expected that stereotype rebound effects (e.g., judgments and behaviours) for the aid-related poor would differ from those reported for previously tested target groups. Stereotype rebound effects for this target group might occur as: a) positive responses, such as approach behaviour or positive judgments, b) occur as negative responses, such as avoidance behaviour or negative judgments, or c) not be evidenced in suppressors’ responses. Four experiments were designed to explore the effects of stereotype suppression, and to examine stereotype rebound effects in perceivers who suppressed (i.e., were asked not to think stereotypically) their stereotypes of the aid-related poor. In the first experiment (N = 29), there was no evidence of stereotype suppression in suppressors’ essays about a poor African youth. In the second experiment (N = 24) however, suppressors sat significantly closer than controls to the alleged seat of an African student (a reverse rebound effect). In the third experiment (N = 35), suppressors again showed evidence of approach behaviour when interacting with African poor individuals, and demonstrated avoidance behaviour when interacting with African wealthy individuals. Experiment 4 (N = 70) used implicit measures of stereotyping; participants were asked to name the ink colour of stereotypic and nonstereotypic trait words presented immediately following two category primes. Suppressors tended to be faster than controls to name the colour of positive trait words and slower than controls to name the colour of negative trait words that were stereotypic of the African poor target group. This difference in response times implies that, for suppressors only, the salient stereotype features being primed were positive rather than negative; thus eliciting a positive rebound effect. Stereotype rebound effects may therefore not always be evidenced by higher levels of negative or prejudiced responding in suppressors, but can also appear as positive evaluations and approach behaviours. Findings from this research (i.e., stereotype rebound effects can be either positive or negative) have important implications for those viewers of aid advertisements who suppress unwanted stereotypes, especially as rebound effects for the aid-related poor are associated with unconscious behavioural and cognitive responses.