The Asian American Vote in 2020: Indicators of Turnout and Vote Choice

We acknowledge that this survey was conducted some months after the 2020 election. In March 2021, a mass shooting that targeted Asian women occurred in Atlanta, Georgia (see Knowles et al., 2021). One concern is that this event and other hate incidents occurring in early 2021 may have influenced how Asian American survey respondents viewed racial discrimination against their group and which would result in a spurious relationship with voter turnout or candidate choice in the election that occurred months prior. However, it is unlikely that the Atlanta shooting dramatically changed Asian American attitudes about discrimination between the election and the collection of this survey. Anti-Asian attitudes had already been on the rise in 2020 and so the Atlanta shooting did not necessarily present “new” information that anti-Asian hate existed in the United States between November 2020 and collection of the survey. Moreover, a separate study by Leung, Kim and Masuoka (2021) found that Asian American attitudes about discrimination were unchanged in response to new events in 2020 when the initial spike of anti-Asian attitudes occurred. This suggests that attitudes about discrimination are not strongly influenced by new events. A different study by Arora et.al. 2021 found that media coverage and attention to the mass shootings in Atlanta were very short lived. Analysis in this study shows that audiences were only responsive to social media posts about the shooting that only lasted for about a month after the event. Taken together, we do not believe the Atlanta shooting is a confounding event in this study that altered survey respondents’ attitudes about racial discrimination.

Data in this analysis are weighted to fall within the margin of error of the adult population in the 2019 Census ACS 1-year data file for age, gender, education, nativity, and ancestry. A post-stratification raking algorithm was used to balance each category within ± 2 percent of the ACS estimates. See https://​cmpsurvey.​org/​2020-survey/​.

We acknowledge that social desirability bias may influence response to this survey item. Unfortunately, a validated turnout variable was not included in the 2020 CMPS initial release, at the time when this article was written. We follow a tradition in political science research which uses self-reported measures of voter turnout (Brady et al., 1995; Dawson 1994; Leighley and Nagler 1992).

For question wording and coding of all variables please see online Appendix A.

We also considered how experiences during the COVID-19 pandemic may have impacted voter turnout and presidential vote choice among Asian Americans. We considered the role of experiencing a COVID-19 infection and experiencing economic hardship during the pandemic on our two dependent variables. This analysis found that these health-related factors were not significantly related to voter turnout nor presidential vote choice. Therefore, we excluded these factors from consideration in this study. We present our analysis of COVID-19- related experiences in the Appendix B, Tables B1-B4.

The CMPS is currently embargoed to collaborators and contributors only. See: https://​cmpsurvey.​org/​2020-survey/​. The survey will be posted for public use on ICPSR sometime in 2023. We have posted code for the analysis on the Political Behavior Dataverse here: https://​doi.​org/​10.​7910/​DVN/​XI7MWK.

This rate is slightly higher compared to estimates documented by other studies, but the difference is not large. For example, in an analysis of vote returns conducted by Catalist, it was estimated that about 62% of eligible Asian American adults voted in the 2020 presidential election (Ghitza and Robinson, 2020). An analysis of 2020 Current Population Survey data finds that approximately 60% of Asian American eligible adults voted in the presidential election (Ramakrishnan 2021). The advantage of using survey data is that we can look at individual-level attitudinal correlates of turnout.

Vietnamese and Filipino Americans may have been the least likely to turnout in 2020 due to under-recruitment in those communities (see Appendix, Table B11). In the 2020 CMPS, we found that these two Asian American national origin groups were those least likely to be contacted by political organizations about voting.

We also ran the Demographics Only Model separately on native-born and then on foreign-born Asian Americans (see Appendix B, Table B5). We find fairly consistent demographic predictors for Asian Americans across immigrant status. Those who were older and with higher socioeconomic status were more likely to turnout to vote in 2020 for both native-born and foreign-born Asian Americans.

We also ran the Demographics Only Model separately on native-born and then on foreign-born respondents (see Appendix Table B8). For native-born Asian Americans, we find that women and those who attended religious service more frequently were more likely to vote for the Democratic ticket in 2020. Among foreign-born Asian Americans, education and religious service attendance were significant predictors of vote choice. National origin differences also emerge across foreign- born Asian Americans; Chinese, Filipino, Japanese, and Vietnamese Americans were less likely than Indian Americans to vote for Biden and Harris.

Regression results for the Full model and Demographics only model for each national origin group is presented in Appendix Tables B9 and B10. We find that for each national origin group, different demographic groups were more likely to vote for Biden. For Filipinos, it was the higher educated and native-born. For Indians, women were strong Biden supporters. For Koreans, younger and higher educated voters supported Biden. For Vietnamese, younger and higher income voters preferred Biden.

We note that media attention to anti-Asian hate incidents against the elderly is distinct from the fact that several sources have found that younger Asian American adults are more likely to report experiencing anti-Asian hate (for example, see: Lee and Ramakrishnan 2021).

Similarly, personal experience of threat has been found to politically mobilize immigrants. See Pantoja, Ramirez and Segura (2001).