Possible Implicit Mechanisms of Minority Representation

Abstract

Minority representation is an important topic for political science—how do members of a racial majority group identify with the political goals of a minority, even when research documents widespread anti-Black bias? Does pro-Black policy support require an individual to be unbiased, or can such support emerge despite internalized anti-Black bias? This paper draws a distinction between two types of implicit racial attitude measures based on functionalist theories of attitudes (Katz, 1960; Smith, Bruner, & White, 1956) and research regarding automatic empathic processes (Decety & Jackson, 2004; Preston & de Waal, 2002). According to this distinction, some attitudes evaluate racial groups as attitude objects (evaluative associations), while others involve automatic identification with them as people (relational associations). I use subliminal racial priming for evaluative associations and a measure of implicit closeness to Blacks (Craemer, 2008) for relational associations. The evaluative measure captures association strength between a racial stimulus and race-unrelated positive or negative target words. The relational measure assesses association strength between the respondent's self-concept and the mental representation of a racial group based on implicit “self-other overlap” (Aron, Aron, Tudor, & Nelson, 1991). Implicit measures are obtained in a well powered online reaction time study (n = 1,341). Online results are evaluated against a representative telephone survey (n = 1,200). Consistent with the extant literature, a significant anti-Black bias emerges in evaluative associations. In contrast, a significant pro-Black effect emerges in relational associations and the two implicit measures are statistically unrelated. Both measures predict pro-Black policy support independently of one another and net of other factors. Implicit closeness to Blacks predicts pro-Black policy support even among White respondents suggesting that minority representation based on relational associations may be possible despite widespread anti-Black evaluative bias.