Keywords: Racial inequality beliefs; Diversity and inclusion; Popular culture; Quasi-experiment

Abstract: This study examines how heightened racial representation shapes beliefs about the causes of racial inequality. To explore this, it focuses on the unprecedented diverse racial representation in the 2016 Tony Awards, a prominent American entertainment award celebrating excellence in Broadway theater. Leveraging the 2016 General Social Survey data in a quasi-experimental setting, the analysis investigates the change in beliefs among respondents interviewed before and after the Tony Awards. The results indicate that individuals are less likely to perceive discrimination as the primary cause of African Americans' social disadvantage after the Awards. This effect is heterogeneous across demographic and social groups, particularly impacting white individuals and Republicans. This study suggests that engagement with cultural products highlighting progress toward diverse racial representation may lead to a temporary underestimation of persistent racial discrimination, even in the presence of enduring inequality.

[ Article ]  [ Replication Package ] 

Keywords: Racial segregation; Ethnic diversity; Population growth; Rank

Abstract: As the United States undergoes demographic shifts where multiple racial and ethnic groups increasingly coexist, understanding how these changing compositions affect residential segregation is crucial. We demonstrate that segregation responds to the rank of racial minority groups by population size---not only their absolute group population sizes. We use U.S. census data from 1990, 2000, and 2010 to show that segregation follows rank-triggered tipping: segregation increases when the relative ranking of minority groups changes. Holding group population shares constant, both across groups within the same county and within the same group over time, a higher rank is associated with higher segregation. Ranks concentrate neighborhood change in specific locations: White flight and minority clustering intensify in census tracts where the higher-ranked group already has a substantial presence. Our findings reveal that residential sorting operates through two complementary mechanisms: traditional share-based tipping (driven by absolute population sizes) and rank-triggered tipping (driven by relative group size). These findings help explain the persistence of segregation across different minority groups and temporal variation in segregation within the same locations, advancing our understanding of how demographic diversity translates into segregation. 

[ Manuscript ] 

* Submitted. Previously titled as "#MeToo and Beyond: Addressing Gender Inequality in the Workplace"

Keywords: Sexual harassment; Workplace; Gender inequality; Social movement

Abstract: Sexual harassment is a key factor perpetuating gender inequality in the workplace, disproportionately affecting women and imposing substantial economic costs. This paper examines whether social movements raising awareness of sexual harassment can influence women's workplace outcomes. Focusing on the #MeToo movement, I exploit temporal variation around its onset and geographic variation in local engagement, measured by Google search intensity across the United States using a difference-in-differences design. I first document an increase in workplace sex crime reporting among female victims in higher-engagement areas, with no corresponding change in incidents outside the workplace, consistent with the movement having shifted the broader workplace environment. I then document downstream economic effects: in higher-engagement areas, women's labor force participation and earnings increase modestly but measurably, with little change in hours worked.  These findings suggest that local engagement raised the visibility and cost of harassment, facilitating gains in women's labor market outcomes.

[ Manuscript ]  [ Poster (September 2025) ] 

Keywords: Boundary; Immigration; National identity; Asian Americans

Abstract: Who belongs to the nation is not fixed but negotiated through the drawing of symbolic boundaries. This paper asks how local immigration shapes these boundaries. It conceptualizes national identity as a relational process in which minorities may claim membership, but shared membership also depends on recognition by the majority. I examine this process in the context of Asian immigration to the United States, where Asians are a fast-growing and socioeconomically integrated group that nonetheless remains perceived as foreign. Using large-scale survey data with direct measures of implicit and explicit national identity for Asian and non-Asian Americans, I employ a shift-share instrumental variable design that exploits initial Asian settlement patterns and national immigration inflows. The results show that increases in the local Asian immigrant population weaken implicit associations between Asians and foreignness among both Asian American and non-Asian respondents. Taken together, the findings shed light on how immigration can reshape racialized understandings of who counts as American.

Draft coming soon!

Keywords: open science; transparency; data sharing; computational reproducibility

Abstract: Public sharing of data and analysis code enables independent verification of published findings, yet systematic evidence on such sharing in sociology remains scarce, even as audits have been conducted in psychology, economics, and political science. We examine 730 empirical articles published between January 2019 and June 2024 in three leading general-interest journals: American Sociological Review, American Journal of Sociology, and Social Forces. For each article, we record whether a publicly accessible replication package is provided, and for those that share materials, we attempt to computationally reproduce the main-text tables and figures. Across the sample, 9.9% of articles provide replication packages, with substantial variation across journals (5.2% to 22.9%) and research types (0% for qualitative studies, 30% for experiments). These rates fall well below those reported for political science and economics. Among 72 packages examined in detail, more than half cannot be verified due to missing or incomplete materials. At the same time, no article is wholly non-reproducible, and 22% are fully or largely reproducible. Sociology is not unusually prone to errors when materials are shared and runnable - it simply shares them far less often. We discuss implications for journal policies and for transparency standards that reflect the field's methodological diversity.

[ Manuscript ]