Attitudes, actions, and constraints toward coronavirus vary by gender, income, and politics

June 3, 2020

New paper from University of Michigan researchers finds different factors at work.

Various demographic groups in the U.S. respond to the COVID-19 pandemic in different ways, according to a new University of Michigan study.

The researchers found substantial gaps in behaviors and beliefs across gender, income and partisanship lines; in constraints across income levels; and in risk tolerance among men and women. The researchers say that officials hoping to control the virus should aim to understand how different factors may drive behavior for specific groups.

“Clearly, in the COVID-19 pandemic, there’s a lot of discrepancy around what people believe to be the truth,” coauthor and Ross School of Business Professor Yeşim Orhun said. “There are also large disparities in the extent to which people have the means to avoid exposure. As governors think about reopening the economy and relaxing restrictions, thinking about how different demographics may react differently and be affected differently is going to be very important.”

The U-M researchers  — Orhun; Professor Ying Fan of the College of Literature, Science, and the Arts; and Ross PhD student Dana Turjeman — surveyed 5,500 adults in April on their beliefs and behavior related to the pandemic. They also examined a database of location information from smartphones.

Their findings so far, as detailed in a new working paper, include:

  • Smartphone geo-location data revealed that U.S. counties with higher shares of women and counties with lower shares of poverty reduced both non-work and work trips compared with otherwise similar counties in the same state. Counties with higher shares of Democrats responded more in reducing non-work trips, but less in limiting work trips.
  • Survey data revealed that women, people with higher incomes, and those identifying as Democrats were more likely to observe social distancing and take hygienic actions such as wearing a mask, washing hands more often, and wiping down groceries. Low-income individuals were less likely to work from home and also less likely to avoid large gatherings and public transportation compared to higher-income respondents.
  • People who are otherwise similar (in, for example, employment, ethnicity, age, and education) and live in similar local communities (in terms of COVID-19 risks) can have considerable gaps in their beliefs regarding the COVID-19 pandemic. Women and Democrats are more pessimistic about their health outcomes and/or chances of being infected, expect more deaths in the U.S., and believe more strongly in the effectiveness of state restrictions regarding health outcomes. They are also more worried about the health and economic well-being of themselves and those around them.
  • While low-income respondents worry more about their own economic well-being compared to high-income ones, they hold more optimistic beliefs about deaths in the U.S. and believe less in the effectiveness of state restrictions.
  • In addition, female respondents and Democrats are less willing to tolerate risk than their counterparts, and low-income respondents are more likely to be constrained in their ability to self-isolate and to work from home.

“We documented substantial differences in actions across the gender line, the partisanship line, and the income line. But the divides across gender and income seem to be at least as strong as, if not stronger than, the divide across political parties in some cases,” Orhun noted.

One key lesson from the study is that explanations for the behavior differences may vary depending on the demographic. For example, the researchers found that the differences between Republicans’ and Democrats’ behaviors are associated with the two groups’ diverging beliefs.

However, constraints are likely to be additional contributors to the differences between lower- and higher-income groups, such as low-wage workers not having the option of working from home. And differences between men and women may be related to a tendency for women to avoid risks, the researchers concluded.

“While all these behavioral differences are substantial, they seem to be associated with different factors,” Orhun said.

Knowing how different factors can affect the behavior of different groups, the researchers advise that keeping all populations in mind will be important for public officials working to control the spread of the virus, even as the economy reopens. The researchers suggest that a unified message is important, but different strategies may be required to reach different demographics. For example, targeted policies may be necessary to get past the challenges facing lower-income individuals.

“As the economy opens up, we should be thinking about different demographics, their innate preferences, and the constraints that they face,” Orhun said.

Yesim Orhun is an associate professor of marketing and a Michael R. and Mary Kay Hallman Fellow at the University of Michigan Ross School of Business. Ying Fan is an associate professor of economics at the U-M College of Literature, Science, and the Arts. Dana Turjeman is a PhD student at the Ross School of Business.

This original article was published as a Ross Thought in Action article.