Aversive Affect Versus Racism as Predictors of Racial Discrimination in Helping

Abstract

We examined aversive affect and racism as predictors of differences in helping White versus Black targets. According to aversive racism theory, Whites may express egalitarian attitudes but experience discomfort in interracial interactions, producing discrimination. Participants completed racism measures and reported their likelihood of helping White or Black targets. Racism negligibly predicted discriminatory helping across studies. In Studies 2 and 3, participants experiencing aversive affect were less likely to help Black than White targets. Results demonstrate negative feelings, more so than racial biases, impacts discriminatory helping. We hope to inspire future research examining why White bystanders experience aversion in interracial helping.

Notes

Notes

¹ Dummy coded 0 = nonspecified race, 1 = Black.

² The MRS, ATB, and RAS measures were standardized and entered into separate regressions.

³ Standardized and treated as a continuous variable.

⁴ The RAS measure demonstrated lower reliability, which may account for the inconsistencies in our findings.

⁵ Dummy coded 0 = nonspecified race, 1 = Black.

⁶ The MRS, ATB, and RAS measures were standardized and entered into separate regressions.

⁷ Dummy coded 0 = Need, 1 = Merit.

⁸ The RAS demonstrated lower reliability, which may account for the inconsistencies in our findings.

⁹ If they provided their contact information, it was sent to the Keep America Beautiful Foundation in Topeka, Kansas.

¹⁰ Dummy coded 0 = White, 1 = Black.

¹¹ Standardized RAS scores.

¹² Dummy coded 0 = White, 1 = Black.

¹³ Standardized RAS scores.

¹⁴ G*Power 3.0.10 (Faul, Erdfelder, Lang, & Buchner, 2007) confirmed the sample size provided a power level greater than 0.99 for the ability to detect a large effect.

¹⁵ Participants were slightly more likely to help (βs = .51–.55) and to provide direct help (βs = .03–.08) in the car scenario than the math scenario.

¹⁶ Participants responded to several items about their perceptions of helping (e.g., perceptions of victim blaming, responsibility to help, costs of helping). However, results revealed that aversive affect, racism, and target race did not influence these variables. These measures and results are not provided in this report; a full report of variables and results is available by request from the first author.

¹⁷ Dummy coded 0 = White, 1 = Black

¹⁸ Standardized

¹⁹ Standardized RAS, MRS, and ATB scores and entered into separate regression analyses

²⁰ Dummy coded 0 = White, 1 = Black.

²¹ Standardized.

²² Standardized MRS, ATB, and RAS scores were entered into separate regression analyses.

²³ 1 = no help; 2 = help.

²⁴ Dummy coded 0 = White, 1 = Black.

²⁵ Standardized.

²⁶ Standardized MRS, ATB, and RAS scores were entered into separate regression analyses.

²⁷ The interactions between race and scores on the RAS (B = −0.30, SE = .37, β = −0.11) produced results that were inconsistent with the pattern demonstrated by the interactions between race and scores on the MRS (B = 0.10, SE = .32, β = 0.04) and ATB (B = 0.13, SE = .32, β = 0.06). Therefore conclusions were drawn from the more consistent pattern demonstrated by racism’s interaction with MRS and ATB scores.

²⁸ The interactions between race and scores on the RAS (B = −0.13, SE = .30, β = -0.06) produced results that were inconsistent with the pattern demonstrated by the interactions between race and scores on the MRS (B = −0.29, SE = .25, β = −0.15) and ATB (B = −0.20, SE = .25, β = −0.11). Therefore we probed only the Race × Racism interactions that included the MRS and ATB.

²⁹ The interactions between race and scores on the RAS (B = 0.10, SE = .25, β = 0.04) produced results that were inconsistent with the pattern demonstrated by the interactions between race and scores on the MRS (B = −0.33, SE = .29, β = −0.11) and ATB (B = −0.57, SE = .29, β = −0.20). Therefore we probed only the Race × Racism interactions that included the MRS and ATB.

³⁰ The interactions between race and scores on the RAS (B = −0.17, SE = .29, β = −0.08) produced results that were inconsistent with the pattern demonstrated by the interactions between race and scores on the MRS (B = −0.41, SE = .24, β = −0.22) and ATB (B = −0.35, SE = .24, β = −0.20). Therefore we probed only the Race × Racism interactions that included the MRS and ATB.

³¹ The interactions between race and scores on the RAS (B = −0.03, SE = .25, β = −0.01) produced results that were inconsistent with the pattern demonstrated by the interactions between race and scores on the MRS (B = −0.57, SE = .28, β = −0.20) and ATB (B = −0.65, SE = .29, β = −0.22). Therefore we probed only the Race × Racism interactions that included the MRS and ATB.

³² To support that participants had lower levels of self-reported prejudice, the average values of racism scores across all studies were examined. For each measure, participants responding using a 1-to-9 scale, where higher scores were indicative of higher racism. Participants’ average MRS scores ranged from 2.43 (SD = 1.15) to 3.31 (SD = 1.26)- average ATB scores ranged from 2.38 (SD = 0.98) to 3.54 (SD = 1.25)- and average RAS scores ranged from 4.04 (SD = 1.63) to 4.91 (SD = 1.44). All of these average values are below the midpoint of the scale, suggesting that, on average, the participants in these studies did not report extremely negative racial prejudices.

³³ MRS αs = .69–.83; ATB αs = .78–.87; RAS αs = .45–.76.