Measuring Work from Home in the Cross Section

Abstract

We compare survey-based measures of work from home (WFH) in the cross section. While the surveys differ in how comprehensively they measure WFH, they are highly correlated in the cross section of US states, suggesting that they will yield similar causal effects. Researchers should carefully consider the trade-off between how comprehensively WFH is measured and measurement error, with the American Community Survey well suited for low levels of aggregation and the Survey of Working Arrangements and Attitudes and Real-Time Population Survey well suited to measuring distinct types of WFH. We also document that the experimental 2020 ACS replicate weights produce errors that are too small.