Attitudes about science were associated with compliance with shelter-in-place policies during the initial stages of the COVID-19 pandemic in the United States, according to research that analyzed anonymous cell phone location data. The study indicates that regions where people are more skeptical of science tend to adhere less strictly to stay-at-home orders. The findings have been published in Nature Human Behaviour.
Following the outbreak of COVID-19, social scientists quickly became interested in studying factors that impact compliance with government policies that mandate physical distancing.
“Since we are from economics and public policy backgrounds, we were naturally interested in studying individual behavior in response to public policy in the context of the COVID-19 pandemic. We got the idea for the paper when state governments across the U.S. started introducing shelter-in-place policies in a staggered fashion in March 2020,” said study author David Van Dijcke, a PhD student at the University of Michigan.
“We realized we could use the variation in the timing of when those policies were introduced to trace out their effects. Since it was apparent to us that non-compliance with the policies — people not staying at home — would be an important issue for their efficacy, we started thinking about what might affect such non-compliance. While we look at science skepticism, other studies have found important roles for partisanship and poverty as well.”
“Around the same time, we stumbled across SafeGraph, the company that provided us with the anonymized mobile device data that we used to estimate the extent to which people were staying home,” Van Dijcke said.
The researchers measured responses to the shelter-in-place policies at the county level by analyzing location data from more than 40 million mobile devices across the United States.
Van Dijcke and his team used data from a previous study on climate change opinions, which aggregated data from 12 nationally representative surveys, to assess science skepticism. The surveys included responses from 12,061 individuals in total and the data were used to estimate the percentage of people per county who agreed with the statement that global warming is caused by humans.
Because of the lack of granular geographic data on science skepticism, “we used belief in anthropogenic (human-made) global warming as a proxy for science skepticism and validated this measure by benchmarking it against measures of science skepticism from other, smaller-scale datasets,” Van Dijcke explained.
Those other datasets included the American Values Survey, which asks respondents the extent to which they agree with the statement “I am worried that science is going too far and is hurting society rather than helping it,” and the World Values Survey, which includes survey items such as “We depend too much on science and not enough on faith.”
The researchers found that the proportion of people who stayed at home after shelter-in-place policies went into effect tended to be higher in counties with lower levels of skepticism compared to counties with higher levels of science skepticism.
Previous research has found that shelter-in-place policies tended to be less effective in regions with a greater share of Donald Trump voters. But Van Dijcke and his colleagues found that their results held even after controlling for political partisanship.
“The main takeaway is that whether or not people stayed at home during the first COVID-19 lockdowns in the States depended to a significant extent on whether they were skeptical about science,” Van Dijcke told PsyPost. “That is the case irrespective of people’s political affiliation, income, education, etc. We also find some evidence that science skepticism undermined compliance with other public health interventions during the pandemic, such as mask-wearing and vaccination. We think these are important findings since they underline the importance of science education and communication, as well as the danger of misinformation about these topics.”
“A caveat to our study is that it applies to the United States during the first wave of the pandemic, and thus may not be generalizable beyond that setting,” Van Dijcke noted. The study examined the proportion of people who stayed at home between March 1 and April 19, 2020.
But the most important limitation of the study is the fact that the researchers had to rely on belief in anthropogenic global warming as their primary measure of science skepticism.
“An obvious lacuna to fill is the availability of large-scale, representative data on science skepticism that can be mapped to granular geographies in the United States such as counties,” Van Dijcke said. “I think the pandemic has forcefully demonstrated how detrimental science skepticism can be to the implementation of public policy. Such data would open the way for a large array of additional questions regarding science skepticism to be studied, since researchers could link it to any other data available at the county level, most prominently Census data.”
However, the results are in line with another study published in Nature Human Behaviour, which found that people with lower levels of trust in doctors, scientists, economists, professors, and experts were less likely to engage in behaviors intended to prevent the spread of COVID-19.
The study, “Science skepticism reduced compliance with COVID-19 shelter-in-place policies in the United States“, was authored by Adam Brzezinski, Valentin Kecht, David Van Dijcke, and Austin L. Wright.
