Crossing the racial line: The fluidity vs. fixedness of racial identity

ABSTRACT

Five studies investigated perceptions of individuals whose identity claims violate societal ascriptions of group membership. Studies 1–3 showed that perceivers dislike targets whose claimed race/ethnicity does not match either of their parent’s racial/ethnic ancestry, delegitimize their identity claims, and deny their claimed identity relative to targets whose claimed race/ethnicity matches at least one of their parents’ racie/ethnicity backgrounds. Study 4 showed that mismatched religious identities are not similarly devalued, suggesting that perceived misrepresentation of racial/ethnic identity holds special significance as a violation of social norms. Study 5 found that racial essentialism was associated with increased disparagement of targets with two White parents who claim a Black identity, but not of targets with two Black parents who claim a White identity.

KEYWORDS:

• Race
• identity claims
• racial essentialism
• social identity
• racial categorization

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.

Notes

1. For the rest of the paper we will refer to a transracial identity claim as a “mismatched” identity claim; i.e., a racial identity claim that does not match the biological ancestry of either parent.

2. Black males are more likely than Black females to be seen as exemplars of racial groups (Eagly & Kite, 1987) and research on intersections of racial and gender identity has demonstrated that non-ambiguous Black female targets are seen as “less” Black than non-ambiguous Black male targets (Goff, Thomas, & Jackson, 2008). These findings guided our decision to use only male targets for our experimental stimuli to limit the impact that gender may have on racial categorization of the target.

3. For Study 1, there were no significant differences in ratings of primary outcome measures based on which face was associated with the target (target liking: t(376) = .108, p = .914, d = .011; identity legitimacy: t(376) = .319, p = .75, d = .032; racial categorization: t(376) = .013, p = .98, d = .006) thus they were collapsed for analyses. For Study 2, there were no significant differences in ratings of primary outcome measures based on which of two ethnically ambiguous facial photos was included in the target profile, (target liking: t(778) = .264, p = .792, d = .018; identity legitimacy: t(777) = .423, p = .677, d = .03; racial categorization t(783) = .362, p = .718, d = .025); thus, they were collapsed for analyses. For Study 5, there were no significant differences in ratings of primary outcome measures based on which face was associated with the target (target liking: t(523) = .135, p = .904, d = .01; identity legitimacy: t(523) = .440, p = .624, d = .033; racial categorization t(523) = .235, p = .814, d = .019); thus they were collapsed for analyses.

4. Faces used as stimuli were displayed against a White background and contained no external hair and/or clothing cues. For Study 1 and 5, Black and White faces were obtained from Minear and Park’s database (2004) and morphed (50%:50%) to create one of two racially ambiguous faces. These pre-tested racially ambiguous faces (see Peery & Bodenhausen, 2008 for detailed description of pretesting process) all had neutral expressions and were matched for age (18–29). For Study 2, faces used for the target profiles were created with the Morpheus software (v 5.55.1) by morphing Latino and White faces (50%:50%) obtained from the same face database as used in Study 1 (Minear & Park, 2004). All faces had neutral expressions and were matched for age (18–29). See OSM for a detailed analyses of how we chose the final faces to be used as stimuli as well as analyses conducted to confirm they could be collapsed for our primary analyses.

5. Study 1: No significant differences emerged depending on whether the target’s claimed identity aligned with his mother’s or father’s racial identity (target liking: t(111) = .298, p = .766, d = .056; identity legitimacy: t(111) = .407, p = .685, d = .076; racial categorization: t(111) = .184, p = .855, d = .029). Study 2: No significant differences emerged depending on whether the target’s claimed identity aligned with his mother’s or father’s racial identity (target liking: t(308) = 1.35, p = .140, d = .153; identity legitimacy: t(307) = .407, p = 1.05, d = .119; racial categorization: t(309) = .326, p = .744, d = .037). Study 3: No significant differences emerged depending on whether the target’s claimed identity aligned with his mother’s or father’s racial identity (target liking: t(218) = .211, p = .833, d = .028; identity legitimacy: t(218) = .982, p = .327, d = .133; racial categorization: t(217) = .149, p = .881, d = .020). Study 4: For the matched-one condition, no significant differences emerged depending on whether the target’s claimed identity aligned with his mother’s or father’s religious identity (target liking: t(126) = .166, p = .868, d = .029; identity legitimacy: t(126) = .316, p = .762, d = .055; racial categorization: t(126) = .542, p = .588, d = .095).

6. This additional paragraph was not included for the matched-both condition.