RAPID: Surveying the American Public in the Time of COVID-19: Comparing Video to Face-to-Face Interviews

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

The National Science Foundation has typically funded face-to-face survey interviews with nationally representative samples of Americans. It has seemed inconceivable that face-to-face interviewing might not be possible. But then in 2020, the COVID-19 (coronavirus) pandemic arrived. States issued stay-at-home orders, required social distancing, and brought most in-person research to a standstill. In light of these circumstances, this project develops a mixed-mode design for surveys that will facilitate rigorous data collection in a period of viral pandemic and public uncertainty. We conduct multiple experiments and simulations that address the feasibility and logistics for the use of remote, live interviewing by video conference. This project builds on a small but powerful mode experiment that shows promising results. A previous experiment allowed within- and between-subject experimental design features to assess differences across modes, with randomization occurring after interviewees have consented to participate in this survey study. While we observed only minimal, statistically insignificant differences in the available data quality metrics—non-differentiation, non-response, don’t know responses, and length of response to open-ended questions—across modes, this may have resulted in part because it was conducted in a controlled environment to maximize internal validity, with all respondents participating in a central on-site location. The experiments and simulations of this project extend the experimental designs to a more externally valid field environment to further test for data quality of video interviewing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.