When the COVID-19 pandemic struck in March of 2020, Northwestern University political scientist James Druckman was exchanging emails with a small group of social scientists from Harvard, Northeastern and Rutgers about how they were navigating COVID’s impact on their work and lives. When California became the first state to lockdown, they decided to collectively embark on a state- and federal-level survey and make data available to the public on a range of topics such as mask-wearing, remote learning, vaccinating and voting by mail.
“We knew there would be national surveys that policymakers and the media would be looking to, and we felt those surveys would be incomplete because there would be important differences across states, both in how people were reacting to what an individual state was doing and then just individual state cultures,” said Druckman, the Payson S. Wild Professor of Political Science in the Weinberg College of Arts and Sciences and associate director of the Institute for Policy Research at Northwestern.
The colleagues found a vendor to distribute the survey within all 50 states and the District of Columbia and received a Rapid Response grant from the National Science Foundation (NSF). Within a few short weeks, they were able to launch the first survey. They published their first survey report on national and state COVID-19 data on April 30, 2020, under the name the COVID States Project.
Since the first survey report published three years ago, the principal investigators have continued to meet weekly on Zoom. Their report findings have been presented to federal and state officials to help inform their decisions and allocate resources more effectively.
Following the release of Report #100 on “Estimating Current Vaccine Rates,” Northwestern Now spoke with Druckman to find out what he has learned since the polling began, and what the future plans are for the COVID States Project.
What themes did you focus on in the surveys?
We tried to consistently ask about a range of COVID behaviors that people were engaging in and support for COVID policies that different states were implementing. We consistently asked about trust in government, trust in science, and trust in specific scientists and political figures; [and] economic hardship because we wanted to see who was experiencing hardship during the lockdowns. From the genesis of the project, it also was important to include a social network questionnaire to see who people were connected to, and how those connections were affecting their mental health, their trust and the behaviors they were engaging in. We used the same questionnaire physicians use to screen for levels of depression and social isolation.
We never anticipated the things that would occur in the country over the next three years. But because we had this methodology down, we were able to pivot in a very quick fashion to incorporate new topics into the survey as they occurred, ranging from George Floyd's murder and the Black Lives Matter protests to the spike in gun purchasing, the 2020 election, Jan. 6, the Dobbs decisions, and the war in Ukraine. For example, we were able to get a survey into the field on Jan. 7 after the Jan. 6 insurrection.
What are some of the biggest takeaways from the surveys?
I see three big takeaways as we enter the post-pandemic period. One is that the political divisions that existed before COVID got larger. Another is that mental and physical health has gotten worse. Third is the gun buying spike and Jan. 6, which are manifestations at the intersection of the pandemic and the political climate.
Because we have such a sizeable amount of data, we can see there is a large number of people who have suffered from long COVID, which doesn’t get a lot of attention, but is very serious. It affects their economic well-being, their psychological well-being, their ability to work and their social relationships.
What concerns you the most about the data?
We’re in a situation where the country, the political system and the health system are under a lot of stress. And it’s unfortunate, because I think, when COVID first hit, some of us including myself, had this sliver of hope that it could bring the country together, because we were all facing the threat of this disease.
In looking at the data trends, one big thing to which I keep coming back is gun purchases. People often buy guns because of feeling diffuse threat — that is threat that they cannot pin down to specific causes. The combination of health, political, racial and economic threat manifested in some sense in gun buying. An estimated 4.5% of guns on the street today were bought after COVID hit, which is an enormous number of guns if you think about the timeframe (Report #37).
Did you achieve your goals?
Yes. The initial goal was to attract a public response to COVID, and we were able to generate a lot of media coverage.
Study reports have been cited in The Wall Street Journal, The New York Times, The Atlantic, The Washington Post, USA Today, CBS News, Nature, NPR and STAT, earning approximately 1,100 media mentions and reaching more than 79 million news readers.
We also were able to provide important data to practitioners and policymakers about media diet, including where vaccine hesitant people are getting information.
What’s next for the project with the end of the pandemic on the horizon?
We’ve been awarded a three-year grant from the NSF to reinvent the project for the post-pandemic era. We're going to keep doing these surveys, but half of every survey will be for researchers who put in an application on a competitive basis. We get a lot of people wanting to use the survey because that allows them to study differences between states, which is hard to do, but we have a mechanism that provides people with the ability to collect these data.
What are the benefits of this multi-university approach?
It's not typical in the social sciences, to have such a team science approach, particularly across universities. But it has been great because we have researchers from four different disciplines on the team, living in three different states who are having very different experiences. The process has been very inclusive and collaborative on co-publishing papers. In addition to the 100 working papers, the team has published more than a dozen journal articles in medical, political science and communication and computer science journals.
We have been fortunate to have excellent students who made it viable. The project has provided fulfilling opportunities for graduate and post-doc students who have gone on to get great jobs.