Drs. Sara Wallace Goodman, Shana Kushner Gadarian, and Thomas Pepinsky received funding from the National Science Foundation to support a survey about Americans’ beliefs and behaviors related to the novel coronavirus. The scholars have already shared early results about the partisan responses to the crisis. APSA staff spoke with Dr. Goodman about the project.
How would you describe the big picture of the project?
We wanted to see people’s attitudes and behaviors as this crisis was emerging. I’m a Europeanist; I study immigration attitudes and civic responsibility. I was in Denmark late February and I saw what was happening there, and what was inevitably coming to the US. I want to give our team a little credit—we were early on the ball. Our application to NSF was in by early-March.
We wanted to know what people were going to do. I wanted to ask about attitudes on immigration policy, shutting down external and internal borders. Tom [Pepinsky] is a political economist, so he wanted to ask about policies and effects on the economy, and Shana [Gadarian] is a political psychologist and wanted to study blame attribution and anxiety. So we pooled all our questions together and developed a survey instrument with a wide range of items. Centrally, we wanted to know if people would respond differently based on partisanship, and how the conditions of deep polarization would affect peoples’ behavior and attitudes.
We found in the first wave that partisanship is the most consistent predictor of mass public responses to the pandemic. We were first out there to identify the strong partisan differences and how they were shaping attitudes towards health, public policy, health behavior.
These ideas were shaped by comparative responses in other countries. We started to see, unlike in Germany and Italy, US political elites releasing contradictory messages where the CDC said one thing, Trump would say another thing, and there wasn’t evidence of bipartisanship. And these are the traditional things that happen during crises! We wondered what the effects will be of this deep polarization, low-trust, noisy information environment.
We wrote NSF for two waves. Our first was fielded at the end of March, right as schools were shuttering, as state orders were being implemented in New York. The second one was in the field in late-April/early-May, after about a month of stay-at-home orders.
You have some early results. Can you talk a little about those?
We agreed we wanted to release whatever we had immediately, for public consumption, to let people know what we were finding as a public good. We found in the first wave that partisanship is the most consistent predictor of mass public responses to the pandemic. We were first out there to identify the strong partisan differences and how they were shaping attitudes towards health, public policy, health behavior. We were able to adjust for a wide range of demographic and geographic differences, so we could control for some of these differences. We observed, on the whole, these strong partisan differences in behaviors and attitudes.
Evidently what needs to happen is for messaging to supersede partisanship. We have some thoughts about how that could happen; but still, you can put a message out there but you can’t open peoples’ ears.
In the second wave, we found a continued pattern of partisan divergence. With the exception of areas where we see the most outbreak or the strongest mitigation policies, partisan differences persist.
What are some early takeaways for policymakers?
Evidently what needs to happen is for messaging to supersede partisanship. We have some thoughts about how that could happen; but still, you can put a message out there but you can’t open peoples’ ears. We haven’t tested any of these ideas yet. What we did have in the original survey was an experiment about valence and partisan endorsements and CDC messaging, and we found no effects.
When we think more generally about the condition of deep polarization and strong partisanship and what happened in terms of national preparation and messaging in February and March, it wasn’t inevitable that peoples’ responses would be partisan. But that was a choice that was made. I recognize the use of passive voice hides all sorts of sins. It’s never inevitable in crises that people respond with partisanship. This is not something you see in Europe. The UK is also a polarized society, but there is no evidence that people responded in their health behavior guided by political identity.
Read more about the project in The Hill.