Partisan segregation may be shaping Asian Americans’ voting behavior

CHAMPAIGN, Ill. — As the 2026 midterm elections draw near, candidates are often focused on currying favor among particular demographic groups. A new study of Asian American ethnic subgroups in the U.S. indicates that political segregation is just one form of spatial isolation that may be shaping Asian Americans’ voting behavior and living environments.

The study’s authors — Ting-Syuan Lin, a doctoral candidate in sociology at the University of Illinois Urbana-Champaign, and Yongjun Zhang, a professor of sociology and affiliate of the Institute for Advanced Computational Science at Stony Brook University — say that while Asian Americans in the U.S. are often considered to be a single racial group, they comprise several ethnic subgroups — including Chinese, Filipino, Japanese and Vietnamese Americans — that differ markedly in their political orientations, socioeconomic resources and migration histories. Lin and Zhang used voter data to explore segregation by race, income and party affiliation among 33 million registered voters in New York and California, with a particular focus on the 2.55 million Asian American voters in these states.

Asian American ethnic subgroups also differ in where they live and the racial, economic and political makeup of their neighborhoods. Yet their differences are often obscured in research that treats them as a single homogeneous group, according to Lin and Zhang, who compared residential segregation among Asian Americans as an aggregate group with that of white, Black and Hispanic voters, and examined differences among six of the Asian American subgroups in the U.S. using individual-level voter data.

Research that focuses on concentrations of Asian Americans in enclaves —neighborhoods with high concentrations of a particular ethnic group — and ethnoburbs, defined as similar communities in suburban areas, limits understanding of how race, income and politics collectively influence these groups’ living experiences, Lin said.

 “Since income and race often line up with political affiliation, political segregation may overlap with racial or income segregation, but they are not necessarily the same thing,” Lin said. “With increasing attention focused on political polarization, and on whether voters choose to relocate to areas with people who share their politics, we wanted to explore these patterns among Asian American subgroups as a first step toward understanding how their residential environments might affect voting behavior.”

For the purposes of the current study, isolation and segregation refer to the extent to which individuals are exposed to neighbors from the same social groups. And, unlike many prior studies, the current study defines neighborhoods through geographic proximity rather than fixed administrative boundaries, allowing a flexible measure of segregation across multiple dimensions that more closely captures voters’ living environments, the researchers said.

Since California and New York are two of the most populous and diverse states, Lin and Zhang based the study on these states’ 2021 voter data — which included residential addresses, party affiliation and demographic information. Lin and Zhang mapped the geographic distribution of Asian American registered voters across the counties in the two states and examined differences among their six largest national-origin Asian subgroups — including Chinese, Indian, Vietnamese, Korean, Japanese and Filipino Americans.

Using geocoded voter locations, Lin and Zhang constructed an individual-level spatial isolation index — a tool that Zhang refined in a previous project — that measured racial, income and partisan isolation for each voter based on their 1,000 nearest neighbors who are registered voters.

When the team examined isolation at aggregate group levels, their initial analyses indicated that racial isolation was highest among white voters, followed by Black, Hispanic and Asian Americans. However, isolation by voters’ party affiliation — identified as Democratic, Republican or other — was highest among Black voters followed by Hispanic voters. Asian American and white voters were nearly identical at the lowest levels, indicating that they resided in more politically mixed neighborhoods, Lin said.

In the second stage of their analysis, Lin and Zhang created multilevel models to assess whether various types of isolation differed among the various Asian subgroups, and they examined how these patterns were associated with individual and neighborhood characteristics.

Lin said they found substantial heterogeneity across the Asian subgroups, with Chinese and Vietnamese voters experiencing the highest levels of racial isolation, followed by Korean and Indian voters. By contrast, Japanese voters exhibited lower racial isolation, and Filipino voters had by far the lowest levels, meaning they were more likely to live near voters of other races, the researchers found. “Even the most racially concentrated Asian subgroups were less segregated compared with white, Black or Hispanic voters,” Lin said.

Japanese voters were the only subgroup found to exhibit substantially lower isolation by income compared with Chinese voters, who were used as the reference group in the study. By contrast, Indian voters in the sample reported the highest levels of isolation by income, and they, Vietnamese and Korean individuals exhibited consistently greater income isolation compared with Chinese voters. Middle-income and high-income voters tended to reside in areas with significantly greater income isolation, compared with lower-income individuals.

 Lin said that in comparison with the racial model, differences in partisan isolation were smaller among the Asian subgroups. Japanese and Filipino voters were found to have slightly higher partisan isolation compared with Chinese voters, while Indian and Vietnamese voters exhibited lower levels. Yet all of these groups were well below the levels of the partisan segregation found among Black voters and Hispanic voters.

When the researchers compared New York voters with their counterparts in California, “in general, isolation overall was higher and partisan isolation in particular was consistently higher among New York voters compared with those in California,” Lin said. “However, in California, Indian Americans exhibited both higher racial and income isolation while Vietnamese Americans experienced higher racial isolation only.”

Although prior studies “mainly talked about political ideology or political segregation as a byproduct of race and income, in our study, we find that it is not that simple,” Lin said. “For example, Japanese Americans have higher incomes and relatively low levels of income segregation, yet they experience relatively high levels of partisan isolation, despite living in racially and economically diverse neighborhoods. And while Vietnamese Americans have higher racial segregation and moderate levels of income segregation, they live in relatively diverse political environments as well.”

“This means that we cannot just use race and income to understand the neighborhoods that people live in. We should also consider other dimensions of their living environments before exploring how these environments might affect their vote,” Lin said.

Zhang’s work was supported by the Office for Research and Innovation — formerly the Office of the Vice President for Research Seed Grant Program — and the IACS Seed Grant Programs at Stony Brook University.