ongoing research
MANUSCRIPTS UNDER REVIEW AND WORKING PAPERS
Who Polices Which Boundaries? How Racial Self-Identification Affects External Classification, with A. Armenta, W. Halm, D. J. Hopkins. (Revised & resubmitted)
This study explores whether Americans agree on the ethnoracial categories that are worth policing. It evaluates how receptive White, Black, Latino, and Asian Americans are to how others self-identify by race/ethnicity. Insights from Bourdieu on classification struggles combined with status characteristics theory and gender research suggest that all Americans will police the higher-status White category more than other ethnoracial categories. Other possibilities include White exceptionalism––only White Americans police the White category most––and ingroup overexclusion––everyone polices their own category most. In a conjoint experiment with two samples we find White, Black, Latino and Asian Americans all police the White category most diligently, i.e., they are less responsive when someone identifies as White than when they identify as Latino, Asian, Middle Eastern or North African, or, in most cases, Black. Our results reveal a consensus across Americans on a racial classification schema that reinscribes the contemporary racial hierarchy.
Calling the Kettle White: How Material Stakes Impact External Racial Classification, with A. Armenta, W. Halm, D. J. Hopkins, G. Sigler. (Under review)
Do people define ethnic/racial categories more narrowly or weigh ascriptive criteria more strongly when material stakes incentivize others to identify in certain ways? Realistic group conflict theory predicts as much. We join this theory with recent research that treats ethnic/racial classification as an outcome. We report the results of conjoint survey experiments with two samples, one of undergraduates from diverse ethnic/racial backgrounds, the other of U.S. resident adults. By randomly varying whether respondents racially classify hypothetical profiles completing a scholarship application or an anonymous survey, we examine under what conditions people are more or less likely to classify others as White, Black, Latino, Asian, MENA or POC. We also test whether classification relies more on ascriptive factors (skin tone or descent) when presumed benefits are involved. Our experiments uncover evidence that respondents rely more on ancestry when deciding whether scholarship candidates are White. However, for other classifications, the substantive differences between the scholarship and survey conditions are small, and respondents from different backgrounds behave similarly. These results are consistent with the basic thesis that stakes matter; they also highlight widely shared concerns about fairness and claiming disadvantage inappropriately.
Who Counts as a "Person of Color"? The Role of Ancestry, Phenotype, Self-Identification, and Other Factors, with A. Armenta, W. Halm, D. J. Hopkins. (Under review)
Use of the term "people of color" has grown. What factors do Americans weigh to classify others as people of color (PoC)? An original, pre-registered conjoint experiment with two samples reveals that ancestry is the strongest predictor of classification as PoC. Self-identified race/ethnicity and phenotype also matter, though not as strongly. The results also refute the notion that all non-White Americans count equally as PoC––profiles who identify as Black or with Black parents are most likely to be classified as PoC and those who identify as Asian or with Asian parents least. Classification works similarly across respondents of different ethnoracial backgrounds and political orientations, revealing a broad consensus around who counts as a "person of color."
Money Whitening in the United States, with S. Donahue, A. Laniyonu. (Working paper)
"Money whitening" refers to the notion that more affluent people are more likely to be classified as White. It has been studied extensively in Latin America. Can money whiten in the US, where race is relatively fixed? Using traffic citation data that includes self- and officer-classified race for millions of motorists, we show that Hispanic-identified motorists are more likely to be classified as non-Hispanic White when driving newer cars. Person fixed-effect models confirm that the same motorist is more likely to be classified as White in a newer car. Further, White officers are more likely to whiten than Hispanic officers, and officers are more likely to whiten as the local socioeconomic gap between Hispanic and White residents grows, findings that support stereotyping as the mechanism through which money whitens. Money whitens in the US, but it only whitens people who are phenotypically ambiguous and whose group is implicated in race-based socioeconomic stereotypes.
Diversity and Prosociality in NYC Neighborhoods: Evidence from a Lost Wallet Experiment, with S. Rieger, D. Baldassarri. (Working paper)
Scholars concerned with the consequences of urbanization and differentiation generally associate ethnoracial diversity with negative social outcomes. But a rich sociological tradition envisions diversification as an antecedent to generalized prosociality. Transcending diversity research's reliance on parochial understandings of prosociality, we field a lost wallet experiment in New York City that addresses methodological flaws in past work. We strategically sample a set of neighborhoods that allow us to disentangle contextual heterogeneity from minority share and socioeconomic status, recovering the association between prosocial behavior and heterogeneity per se. Further, we experimentally manipulate wallet owners' racial and socioeconomic backgrounds, shedding light on the extent to which prosociality in large cities is parochial or generalized. We find that once ethnoracial composition and socioeconomic status are properly accounted for, wallet return rates are unrelated to diversity and negatively associated with economic deprivation. Return rates are comparable across recipients from different racial and economic backgrounds, suggesting that prosociality in NYC today generalizes beyond similar others.