Since the coronavirus pandemic began, Asian, Asian American and Pacific Islander communities have endured a spike in hate crimes, with elderly people attacked on the street and an Atlanta gunman killing eight people, six of them women of Asian descent. Many leaders, including President Biden, have suggested that racist rhetoric used to describe the pandemic helped prompt such attacks. Further highlighting the potential link to the coronavirus, the closely divided Senate recently voted by a 92-6 margin to consider the Covid-19 Hate Crimes Act.

Have the events of the last year affected Americans’ awareness of anti-Asian bias fueled by the pandemic?

Not as much as we might expect. Our research found that awareness of anti-Asian bias actually declined between June and October 2020, and only returned to previous levels in late March 2021. This awareness varied by racial and ethnic group, with members of the Asian community particularly attuned to the rising threat of anti-Asian bias related to the pandemic, and White respondents much less aware.

How we did our research

Working with the research organization NORC at the University of Chicago, we fielded a nationally representative study that asked a total of over 4,000 U.S. adults whether they thought Asians in the United States had “experienced any bias or mistreatment related to the coronavirus” (yes or no) at four different points: in June 2020, late July/early August 2020, October 2020, and in late March 2021, soon after the March 16 Atlanta shootings. Because the March survey took place from March 25 to 29, we were able to examine whether prominent media coverage of violence against Asians increased public awareness.

Data were weighted to national census benchmarks and balanced by gender, age, education, race/ethnicity and region to make our findings more generalizable to the U.S. public. Starting with the second survey, the words “or not” were added at the end of the survey question, but as we explain below, we do not find evidence that these two words influenced responses.

What we found

Back in June 2020, 56 percent of all respondents said Asians had experienced bias due to the pandemic. Just over one month later, in late July/early August 2020, this figure had fallen to 51 percent and remained around there, dipping to 49 percent in October 2020, before increasing to 59 percent in March 2021, right after the shootings.

We don’t think this variation shows deep changes in underlying attitudes. Instead we suspect that this U-shaped pattern reflects the timing of prominent media discussion of anti-Asian bias. President Donald Trump’s use of racist language to describe the virus was heavily covered early in the pandemic and again in June 2020 when Trump used of the phrase “kung flu” during a campaign rally just before our June survey; and as we mentioned, the March survey ran soon after the Atlanta killings.

In other words, awareness of bias has not steadily increased throughout the pandemic. The percentages for June 2020 and March 2021 — 56 percent and 59 percent — are statistically indistinguishable. This suggests that, since June, these acts of violence haven’t produced a net increase on whether Americans think that the coronavirus has prompted bias and mistreatment toward Asians.

Views vary by race and ethnicity

This awareness varied by race and ethnicity, as you can see in the figure below.

White respondents showed the U-shaped dip in awareness, with a majority recognizing coronavirus-related bias in June 2020 and March 2021, but not in between. Not surprisingly, Asian respondents became increasingly conscious of the same bias through October, starting at 65 percent and going up to 83 percent. While Blacks and Hispanics had a dip in consciousness of such bias when it wasn’t as prominent in the news, it wasn’t as deep as that of Whites. And even though the sample sizes for Asians, Blacks and Hispanics are somewhat small, it is notable that these groups acknowledged more anti-Asian bias than Whites in every survey.

Whites also showed the largest increase from October to March of any group, from 41 percent to 55 percent, probably prompted by the Atlanta shootings. While awareness of bias among Asian respondents did not increase from October to March, it remained understandably high; the decrease is not statistically significant.

What does this mean?

Why didn’t awareness of anti-Asian bias due to the coronavirus increase for all groups throughout the last year? While it’s possible that the absence of “or not” from the June survey question affected our results, the consistently lower consciousness of anti-Asian bias among Whites than among Hispanics, Blacks, and especially Asians, as well as the increase in Asians’ awareness of bias and mistreatment after the first survey, point to other explanations. For example, it could be that those who felt more personally vulnerable to anti-Asian bias followed this issue more closely.

It’s also possible that some portion of the public didn’t acknowledge this bias because they assume Asians generally face little discrimination in society. Still others may have recognized the growing violence but responded no to our survey questions about mistreatment related to the coronavirus because they felt it instead stemmed from preexisting anti-Asian racism in the United States.

If recent violence reflects deep-seated racism more than a reaction to the coronavirus, it will take much more than the end of the pandemic, by itself, to bring respect.

Jonathon P. Schuldt (@JonathonSchuldt) is associate professor of communication and member of the Roper Center Board of Directors at Cornell University.

Peter K. Enns (@pete_enns) is professor of government, executive director of the Roper Center for Public Opinion Research, co-director of the Cornell Center for Social Sciences at Cornell University and co-founder of Reality Check Insights.

Katherine Zaslavsky (@KAAZsociology) is a PhD student in the department of sociology at Cornell University.

Byungdoo Kim (@ByungdooKim1) is a PhD student in the department of communication at Cornell University.

This project was made possible by funding from the National Science Foundation (Award number 2029498).