Effects of belief versus experiential discrimination on race-based linked fate

ABSTRACT

Since Michael Dawson first put forward the theory that being targets of racial discrimination increases perceptions of linked fate amongst African Americans, applications of his theory to non-Black and contemporary African Americans have arrived at mixed findings. These inconsistencies may be attributed to different conceptualizations of discrimination and a partiality for personal experiences rather than beliefs about discrimination. Using the 2016 Collaborative Multiracial Post-election Survey, we test the argument both beliefs and experiences with discrimination can affect racial attitudes by comparing how the relationship between beliefs/experiential discrimination and race-based linked fate differs across Black, Latino, Asian, and White Americans. The findings suggest even if non-Whites do not personally experience discrimination, they have beliefs about discrimination towards their racial group, and these beliefs are equally important predictors of their sense of race-based linked fate.

KEYWORDS:

• Linked fate
• discrimination
• stigmatized outgroups
• commonality

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 As shown in Figure 1, about $10\text{\%}$ of respondents replied “don't know.” Findings presented below are robust to dropping these responses.

2 We do however, make an exception for English language interview. A negligible number of Whites and Blacks (1 and 2, respectively) took the interview in a non-English language, so we omitted this variable from the analysis for Whites and Blacks. Although there is obvious ethnic heterogeneity amongst pan-ethnic groups such as Latinos and Asian Americans, we do not control for ethnicity in the primary analysis. In models not reported, most indicator variables for different ethnic groups were not strongly related to the outcome measures. Inclusion or exclusion of them do not change substantive findings reported below.

3 In all models survey weights were applied.