The COVID States Project: Thoughts and Actions in the Pandemic
by Mick O'Leary
Data on the COVID-19 pandemic are ubiquitous. Media of all sorts offer front-page, continually updated reporting, especially on the big five COVID metrics: cases, hospitalizations, ICU occupancies, deaths, and vaccinations. But the pandemic is also enmeshed in complex contexts of public opinion and behavior, which not only shape responses to it, but also affect its progress.
The COVID States Project
The COVID States Project (covidstates.org) conducts large-scale surveys on
U.S. public opinion and behavior relating to COVID-19. New surveys are done
frequently to reflect the changing status of the pandemic.
Tracking these correlates is the work of the COVID States Project, a multi-university partnership that conducts regular large-scale surveys on U.S. public opinion and behavior on key aspects of the pandemic. The results are presented in a series of reports and dashboards, with detailed data and analysis broken down by demographic indicators and at national and state levels. The project is conducted by research units at Northeastern, Harvard, Rutgers, and Northwestern universities. It was launched in March 2020 and is funded by a National Science Foundation grant and other foundation support.
The COVID States Project conducts frequent surveys on large samples of voting-age American citizens. Each survey typically reaches 20,000–25,000 respondents, with representation from each state and the District of Columbia. The survey samples are selected to represent the demographic makeup of the state. An important survey tool is a “feeling thermometer,” which measures the respondents’ overall level of “warmth” (or approval) toward a person, group, or idea.
A COVID States Project report covers an individual aspect or event of the pandemic. The survey results are broken down by highly relevant and informative demographic indicators, including political party, 2020 presidential vote choice, vaccination status, education, gender, urban/rural residence, race and ethnicity, age, and income. Data may also be presented at national and state levels. Each 20–40 page report has numerous bar charts, with accompanying analysis of individual indicators and an overall summary analysis with key takeaways.
There are useful dashboards on key aspects of the pandemic and additional data compilations, research studies, and commentaries. All of the reports and these other collections are on the COVID States Project site.
COVID: WHAT PEOPLE THINK AND DO
The COVID States Project has published more than 5 dozen reports since April 2020. Many examine public opinion on varied COVID issues, including the performance of elected leaders, public health messaging, safety measures (masking and social distancing), shutdowns, and school openings. Others investigate public behaviors, including masking, vaccination choices, social interactions, and travel practices.
The breadth of topics covered by the reports demonstrates how widely COVID concerns have come to influence public consideration of other issues. For example, a January 2021 report on gun purchases shows a massive surge in March 2020, as the full import of the pandemic became widely known. A May 2020 report on voting by mail describes correlations between voting preferences and COVID concerns. The COVID States Project is keenly alert to the rapidly changing issues and debates of the COVID era, devising timely surveys that delve into each new stage: COVID-19 and protests (August 2020), school reopenings (August 2020), holiday gatherings (January 2021), vaccine passports (May 2021), and vaccine resistance among parents (July 2021).
With dozens of large surveys on a wide range of topics among diverse respondents, broad conclusions are elusive. Predictably, major differences in opinions and behaviors often correlate with the respondents’ political standing, as reflected by indicators for political party and 2020 presidential vote. But this pattern is not monolithic. Differences among the other indicators (income, gender, etc.) are large on some topics and small on others. Among the states, COVID opinions and behaviors generally reflect red/blue political orientations. But differences between urban and rural respondents also occur. In other words, the opinion and behavior contexts of the pandemic do not support simple conclusions.
Survey data are aggregated into three trackers that present opinions and behaviors over time. The COVID-19 Tweet Dashboard is an interactive tracker that investigates media messaging on COVID topics, based on COVID-related tweets. It explores the top links, domains, and keywords extracted from 29 million tweets related to COVID shared by more than half a million Americans. The study extracted two tweet groups: the 22 Most Shared News Sites on Twitter and the 23 Most Shared Fake News Sites on Twitter (based on a 2016 study co-authored by one of the COVID States Project’s principal investigators).
The Most Shared News Sites on Twitter group is very much larger than the Most Shared Fake News Sites on Twitter group. The Most Shared News Sites on Twitter is dominated by mainstream news sources, with The New York Times, The Washington Post, and CNN consistently the most shared. In this group, there is only one site from the fake news list: The Gateway Pundit. It is also the runaway leader among the fake news sites.
Among demographic groups, people 65 and older did much more COVID tweet-sharing than younger folks. Older Republicans shared proportionally more URLs from fake news domains than Democrats or Independents. The dashboard used data from January 2020 to October 2020 and is being continued by the Lazer Lab, which is a standalone project by one of the COVID States Project’s principal investigators. This topic is also covered in an October 2020 report, “COVID-19 Fake News on Twitter.”
The State Behavior Dashboard tracks a large set of pandemic-related opinions and behaviors for the U.S. and all states and the District of Columbia from April 2020 to June 2021. The results very roughly reflect the red/blue status of each state.
The Trust Tracker documents which organizations and individuals are judged as handling the pandemic the best. Three respondent categories—Democrats, Republicans, and Independents—evaluated 15 agents, including health providers, government actors (presidents Biden and Trump, among others), and the media. There was some agreement among all three respondent groups: All rated Doctors and Hospitals as well as Scientists and Researchers very highly and Social Media very poorly. Disagreements were predictably along party lines. The biggest monthly change occurred in trust levels of the White House; after the 2020 election, Republicans’ trust in the White House plummeted, while that of Democrats and Independents soared. The Trust Tracker covers April 2020 to April 2021.
THE FULL COVID PICTURE
The topics covered by the COVID States Project reports and dashboards are of course not overlooked by other media, which regularly report on trends in COVID-related public opinion and behavior. Frequently, though, the reporting from esteemed sources such as The New York Times, NPR, The Wall Street Journal, The Washington Post, and WIRED is based on the COVID States Project’s work. Its site has a selected list of almost 3 dozen COVID States Project-derived articles, from these and other prominent news sources. These borrowings fortunately allow the project’s findings, even summarily, to be distributed to very wide audiences. But for the full picture of what people think and do about the pandemic, the COVID States Project is the go-to source.