| Popis: |
Research suggests that anti-Black bias among White Americans is persistent, pervasive, and has powerful negative effects on the lives of both Black and White Americans. Research also suggests that intergroup contact in workplaces can reduce bias. We seek to address two limitations in prior research. First, the workplaces reviewed in prior studies may not be typical. Second, previously observed relationships between workplace contact and bias may stem from selection bias—namely, that White individuals who tend to work with Black individuals are systematically different from those who do not, and those systematic differences explain lower bias levels. To address these issues, we review records (N = 3,359) of White, non-Hispanic, working adults in the geocoded General Social Survey (a nationally representative survey) to examine the relationship between contact and bias after adjusting for an exhaustive set of potential confounders. Using Propensity Score Matching (PSM), we compare individuals who worked with Black individuals with their “virtual twins”—individuals who had the same propensity of working with Black individuals but did not. This approach ensures “apples to apples” comparisons on included covariates and does not rest on functional form assumptions. Via PSM, we estimate that having a Black coworker causes a statistically significant reduction in bias for White, non-Hispanic adults. |
