Assessing the Effects of COVID-19 on Housing Vacancy Survey Estimates Using a Revised Nonresponse Adjustment Factor

Jonathan Spader
Daniel Truver
Peter Mateyka
Patricia Holley
Robert Callis
U.S. Census Bureau

Rising COVID-19 case counts in early 2020 led to changes in the data collection procedures used for the Current Population Survey/Housing Vacancy Survey (CPS/HVS), an important source of information about vacancy rates and the homeownership rate in the United States. This report examines the implications of these data collection changes for CPS/HVS estimates. The analyses draw on multiple auxiliary data sources to understand the extent to which changes in nonresponse outcomes accompanied the changes in data collection procedures. The report then develops an alternative nonresponse adjustment factor that corrects for the observed changes in nonresponse. The results suggest that changes in nonresponse likely contributed to the sharp increase in the homeownership rate estimate for the second quarter of 2020. Conversely, the vacancy rate estimates are not similarly sensitive to the alternative nonresponse weighting adjustment; however, the results illustrate the potential for the vacancy rate estimates to underestimate the actual vacancy levels due to the weighting methodology’s assumption that all nonresponding housing units are occupied. These results suggest that the CPS/ HVS estimates of vacancy rates and the homeownership rate should be interpreted with caution for the period affected by the changes in data collection procedures.