Book Manuscript
Colored by Context: Biracial Political Behavior and the Future of Race in America. Manuscript in advanced stages; proposal in preparation
This project updates our theoretical understanding of how racial cues shape behavior when racial group membership is conditional rather than fixed. I propose “relative salience theory,” which argues that when racial boundaries are flexible—individuals can interact as ingroup members across two racial groups—racial cues exert a larger influence on behavior because they can impact which racial group voters perceive as their ingroup, rather than simply how strongly they identify with a fixed ingroup. Because racial boundaries are flexible for Biracials, their political behavior is largely structured by the extent racial cues render one counterpart group relatively more salient than the other.
Empirically, I test this theory across survey, administrative, and experimental data and use a wide variety of methods and identification strategies. I operationalize racial cues primarily as those supplied by social context—including geographic exposure to different racial groups during early-life socialization and across subsequent residential moves—as well as through major racialized events, such as the murder of George Floyd and ensuing protests, and through experimental priming. I show that these cues profoundly impact Biracials’ racial identities, partisanship, policy preferences, and candidate evaluations—substantially more so than among monoracial voters. I lastly show this enhanced responsiveness is conditional on boundaries being flexible, or the extent Biracials can perceive themselves or be perceived by others as a member of different racial groups.
Selected Works in Pipeline
Gregory John Leslie “A Contextual Theory of Biracial Partisanship” R&R at AJPS
Theories explaining how racial identity and group-based cues shape behavior generally treat race as singular and fixed. I propose a theory of Biracials’ partisanship relaxing this assumption, arguing group-based cues matter more when racial boundaries are permeable. Flexibility to interact as ingroup members in two groups extends the range of social contextual cues Biracials are receptive to, facilitating alignment with the counterpart group most salient in their local environment. To test this, I introduce novel methodological strategies for identifying Biracials in administrative data, measuring neighborhood racial composition, and addressing residential self-selection bias. I show Black-White Biracials are profoundly more likely to register as Democrats (Republicans) when they grow up in Black (White) neighborhoods, and adopt partisan identities nearly identical to the contextually dominant group. These results complicate long-standing assumptions about Biracials’ political solidarity with minorities and clarify how political forces shape behavior when group membership is dynamic rather than fixed.
Gregory John Leslie “Partisan Acquisition Across Racial Groups” Under review
Theories of partisan acquisition are fractured along racial lines. Nearly a century of research argues parents dominate partisan learning among Whites, while minority partisanship theories emphasize contextual and identity-based sources. This study revisits dominant accounts, arguing the same underlying dynamics govern partisan acquisition across racial groups, and that apparent differences largely reflect “when” partisanship is measured across the life course. Using voter files from L2, North Carolina, and Florida, I link first-time registrants to their parents and follow voters as they leave home and encounter new environments. The same pattern emerges across racial groups. Parents’ partisanship is the strongest predictor of party registration but declines after adolescence. Contextual influences are muted during adolescence but grow four- to eightfold in early adulthood after residential independence. Panel-based tests show these patterns are not reducible to residential sorting. Together, these findings reconcile competing theories and advance a more general framework of partisan acquisition.
Gregory John Leslie “The Racial Frontier: Biracials, Machine Learning, and the Future of Racial Group Boundaries.” Working Paper
Researchers often examine whether Biracials are more similar to either their lower or higher status racial groups to make inferences about the racial hierarchy. The former circumstance implies stable racial group boundaries via hypodescent, while the latter augurs assimilation for racial minorities and waning intergroup prejudice. This study develops machine learning algorithms using data on voters’ political characteristics to predict single-race ancestry. Algorithms are then applied to Biracials to determine how likely they are to be classified with either their lower or higher status racial group in order to empirically pinpoint their theoretical placement between them. Leveraging symbolic politics and impressionable-years hypotheses positing stable racial affect beyond early-adulthood, the algorithms are applied separately to generational cohorts to examine whether racial boundaries might be shifting overtime. Although Black ancestry continues to possess the most gravity, overtime all Biracial subgroups are becoming increasingly similar to their higher status racial groups, indicating decreasing adherence to hypodescent in their politics.
Rachel Gillum, Gregory Leslie, and Cara Wong “AI, Race, and Politics” Forthcoming at APSA Presidential Task Force Report, Cambridge University Press
Artificial intelligence systems are reshaping racial and ethnic power dynamics across politics, governance, and scholarly inquiry, yet political science lacks systematic frameworks for analyzing when, where, and through what mechanisms these effects occur. This chapter surveys knowledge across three areas: (1) government use of AI in service delivery, surveillance, and coercive administration; (2) AI's impact on political information environments, mobilization, and electoral administration; and (3) AI as research infrastructure in the production of political knowledge. The methodology section examines how AI tools can introduce systematic distortions that standard disclosure practices do not address. In response, the chapter proposes an AI Measurement Statement (AIMS), a disclosure framework informed by practices in machine learning research and industry, designed to surface group-differentiated measurement risks and support transparency, construct validity, and cumulative knowledge production across political science subfields.
Gregory John Leslie and Natalie Masuoka “Discrimination in the 21st Century” How Can Civil Rights Policies Best Embrace the Growing Mixed-Race Population? In Press in Cambridge Handbook of Multiracial Youth and Families
This project explores how the rapid growth of the mixed-race population raises important questions about whether existing civil rights policies adequately address the experiences of mixed-race Americans. This chapter examines two central challenges: how mixed-race populations should be defined and identified for purposes of policy implementation, and whether the discrimination and social barriers they experience stem primarily from their status as mixed race or from their specific racial ancestries. We review three approaches to identifying mixed-race populations—self-identification, parentage and ancestry, and racial classification by others—and draw on existing research and original analyses of socioeconomic outcomes and experiences with discrimination. We show substantial variation across mixed-race subgroups, suggesting that their experiences and life chances remain closely tied to their particular racial ancestry combinations rather than reflecting a single, coherent mixed-race experience. We conclude by outlining recommendations for adapting civil rights policy to a growing and increasingly complex mixed-race population.
Publications
Christopher T. Stout, Davin Phoenix, Gregory John Leslie, and Elizabeth Schroeder “Emotional Representation: Identifying the Characteristics and Consequences of Elected Officials Mirroring the Emotions of Their Constituents” Political Psychology.
This study examines which elected officials are most likely to mirror their constituents' emotions in public outreach—a concept we term emotional representation. We also analyze the significance of emotional representation for the targeted group. To accomplish these goals, we examine the degree to which members of Congress mirrored Black people's documented increase in expressions of anger following the murder of George Floyd in the summer of 2020. Using a regression discontinuity design and the sentiment analysis of 305,358 tweets, 190,192 Facebook Posts, and 35,409 press releases, we show that descriptive representatives provide the highest levels of emotional representation. To examine the impact of emotional representation, we deployed a two-stage experiment to 390 Black respondents. We find that Black people who increased in anger after being primed with images of police violence view elected officials who engage in emotional representation as more favorable, empathetic, and trustworthy.
Gregory John Leslie, Tye Rush, Jonathan E. Collins, and Matt Barreto (2023) “Perceived Racial Efficacy and Voter Engagement Among African-Americans” Politics, Groups, and Identities [pdf]
This paper seeks to provide clarity on the link between Black people’s perceptions of racism in American institutions and society, and voter engagement. We hypothesize that Black people with high feelings of racial efficacy—the belief that their racial group has equal influence as compared to other racial groups in politics and society—the more likely they will be to vote. Conversely, Black people with low racial efficacy are discouraged from voting. However, confidence in in-group leaders and movements can counteract the demobilizing effect of low racial efficacy. Our analysis uses data from an African American Research Collaborative national survey of 1,200 Black voters. Analysis of turnout in 2016 finds that Black people with low racial efficacy were significantly less likely to vote, all else being equal. However, we find that having highly favorable attitudes toward Barack Obama, Black elected officials, and Black Lives Matter recovers the propensity to vote for low racial efficacy Black people back to levels comparable with their racially optimistic peers. Finally, we conclude with a survey experiment which demonstrates that priming Black people with a low racial efficacy environment results in reduced faith in government, in part corroborating the causal direction theorized in our study.
Gregory John Leslie, Natalie Masuoka, Sarah E. Gaither, Jessica D. Remedios, A. Chyei Vinluan (2022) “Voter Evaluations of Biracial-Identified Political Candidates.” Social Sciences[pdf]
Research shows that, all else equal, voters tend to prefer same-race candidates. However, the increasing prominence of political officials with mixed-racial heritage charges social scientists to reevaluate how voters make race-based evaluations. This study seeks to explore how potential voters—both multiracials and monoracials—make judgements about political candidates with mixed-racial heritage but who vary in the specific racial labels they use to describe themselves. We draw upon an interdisciplinary synthesis of existing literature to inform our theoretical expectations. Namely, we expect that multiracial voters will be most supportive of candidates who signal a common ingroup identity by identifying specifically as multiracial. On the other hand, monoracial minorities and Whites should be more likely to adhere to the one drop rule or hypodescent in their evaluations, meaning they should punish multiracial candidates who identify as White, and reward those who identify with their monoracial minority group. Using an original embedded survey experiment design, we find that multiracial respondents are indeed most supportive of candidates who identify as biracial. In contrast, we find that monoracial minorities and Whites tend to completely overlook racial identity cues, and instead focus on the description of the candidate’s racial heritage. This study adds to a burgeoning literature on attitudes toward multiracial candidates and demonstrates that multiracials and monoracials differ sharply in how they perceive race.
Serena Does, Gregory John Leslie, Ariana N. Bell, Katsumi Yamaguchi- Pedroza, Khadija M. Shalbi, and Margaret Shih “How Racial Miscategorization Harms Multiracial Individuals’ Wellbeing: Comparing Samples in the Continental U.S. vs. Hawai’i.” (2021) Cultural Diversity and Ethnic Minority Psychology [pdf]
The current work explores the effects of racial miscategorization (incongruence between other people’s racial categorization of an individual and that individual’s racial self-identification) and subjective wellbeing of multiracial individuals in Hawai’i versus the Continental U.S.. The study consists of interviews with 55 multiracial undergraduate and graduate students conducted in Hawai’i and the Continental U.S.. Thematic analysis identified two central themes relevant to subjective wellbeing: (i) racial miscategorization and its consequences, and (ii) contextual differences in experiences of miscategorization. Results suggest that racial miscategorization is a pervasive experience among multiracial people and is associated with negative psychological wellbeing. We also found that environments with greater representation of multiracial individuals, such as Hawai’i, are associated with less racial miscategorization, more inclusion, and better psychological wellbeing among multiracial individuals. Theoretical and practical implications for social identity research are discussed.
Gregory John Leslie and David O. Sears “The Heaviest Drop of Blood: The Persistence of Black Exceptionalism Among Multiracials” (2022) Political Psychology [pdf][appendix]
We leverage the emerging multiracial population to reexamine prominent theories of the American color line. A Black exceptionalism hypothesis suggests that part-Black heritage will be more restrictive of assimilation prospects than part-Asian or part-Latino heritage. Black exceptionalism better explains their sorting into the racial hierarchy than does classic assimilation theory or a people of color hypothesis. In the American Community Survey, part-Black heritage dominates subjective racial self-identification among biracial adults and identity assignments to children of interracial marriages. In the 2015 Pew Survey of Multiracials, Black-White biracials’ social identity, social networks, perceptions of discrimination, and political attitudes relevant to race resemble those of monoracial Blacks, whereas Latino-Whites and Asian-Whites are more similar to monoracial Whites than to their minority group counterparts. While part-Latino and part-Asian individuals exhibit some propensity to assimilate into the American mainstream, part-Black Americans are still uniquely situated behind a most impermeable color line.
Gregory John Leslie, Christopher T. Stout, and Naomi Tolbert “The Ben Carson Effect: Do Voters Prefer Racialized or Deracialized Black Candidates” (2019) Social Science Research [pdf]
A plethora of research has explored how blacks and whites respond to deracialized and racialized outreach. However, these studies overwhelmingly focus on individuals' reactions to liberal black elites. We explore whether whites and/or blacks favor co-racial elites who take a conservative deracialized position in the form of support for privatizing social security or a conservative racialized position in the form of advocating for ending the norm of political correctness. Using an online experiment with an oversample of black respondents, we find that whites, and in particular white Republicans, have a strong preference for racialized black conservatives over deracialized black conservatives. In contrast, we find that blacks display a preference for deracialized co-racial conservatives, but view black and white racialized conservatives as being equally likeable.