Introduction

In the early days of the COVID-19 pandemic, American housing, food, and job security– critical elements of public health – were disrupted, but the scale and nature of these shocks were unknown. Policymakers and public health authorities lacked real-time data to respond to the rapidly-moving crisis and were unable to design or lead evidence-informed responses. In the early days of the pandemic in the United States, the Data Foundation and partners observed a gap in the information needed to provide an accurate picture of the prevalence of the coronavirus as well as its impacts on the American people and our society. The COVID Impact Survey (CIS) was an innovative solution to tackle the widespread lack of data on the American public’s experiences and health outcomes during the COVID-19 pandemic. 

Between April and June 2020, the CIS asked households in a nationally-representative sample – as well as subsamples in 18 geographic regions of interest – about their physical health, mental well-being, and economic stability. The survey set a precedent for the government’s eventual establishment of its own real-time pulse surveys used to understand COVID-19’s impacts at scale and over time. 

The Data Foundation’s CIS is a valuable use case about how non-governmental entities in partnership with stakeholders, communities, and the public sector can implement rapid, targeted, collaborative, and detailed data collection strategies and theirimpact on federal data collection strategies. CIS delivered key information quickly, which was actionable and urgently needed at the time. Importantly, CIS was published and available as an open-access resource, along with conditional data that may be requested under agreement so it continues to serve researchers across sectors and disciplines ranging from public health to economics.

A product of a collaboration among data-focused non-governmental entities, the CIS was initiated by an idea proposed by a staff member at the Federal Reserve Bank of Minneapolis, who joined the Data Foundation as a Senior Fellow, Abbie Wozniak. The idea gained steam rapidly through conversations with the Data Foundation President and philanthropy in March 2020. When launched, the Data Foundation formulated a senior leadership team of advisors that included a former chief statistician of the United States, a former director of the National Center for Health Statistics, a former deputy assistant secretary for health at the Department for Health and Human Services, in addition to the key members of the research team. 

With initial funding committed by the David and Lucile Packard Foundation, the survey design, advisor support and other aspects of the project were able to rapidly proceed and develop beginning in April 2020. 

The data collection plan initiated with an alignment of interests on questions from a cohort of nearly 30 economists, sociologists, evaluators, and public health experts who helped quickly refine priority questions through rapid and extensive consultation. The National Opinion Research Center (NORC) at the University of Chicago conducted the surveys under contract to the Data Foundation and also provided advice on questions, as did the advisory panel coordinated by the Data Foundation. The CIS questionnaire was intended to align, to the extent possible, with other existing federal government questionnaires so many questions did not need new cognitive testing, for example. Similarly, the methods and design of the survey were chosen to support traditional survey methods in the 18 regional subsamples and use an established national panel to bolster the validity and reliability of results during a time of uncertainty for much of the American population. Thanks to the range of expertise involved in the development of the survey and its emphasis on efficiency, accuracy, and accessibility, the CIS provides a strong example for future real-time data collection in public health crises and beyond.

Author

Acknowledgements

The Data Foundation would like to thank the Robert Wood Johnson Foundation for its generous contribution which supported this work.

Disclaimer 

This paper is a product of the Data Foundation, with funding from the Robert Wood Johnson Foundation. The findings and conclusions do not necessarily reflect the views or opinions of the Data Foundation, its funders and sponsors, or its board of directors.

Copyright © 2024 Data Foundation. All rights reserved.