Non-response in the American Time Use Survey: Who Is Missing from the Data and How Much Does It Matter?

This paper examines non-response in a large government survey. The response rate for the American Time Use Survey (ATUS) has been below 60 percent for the first two years of its existence, raising questions about whether the results can be generalized to the target population. The paper begins with an analysis of the types of non-response encountered in the ATUS. We find that non-contact accounts for roughly 60 percent of ATUS non-response, with refusals accounting for roughly 40 percent. Next, we examine two hypotheses about the causes of this non-response. We find little support for the hypothesis that busy people are less likely to respond to the ATUS, but considerable support for the hypothesis that people who are weakly integrated into their communities are less likely to respond, mostly because they are less likely to be contacted. Finally, we compare aggregate estimates of time use calculated using the ATUS base weights without any adjustment for non-response to estimates calculated using the ATUS final weights with a non-response adjustment and to estimates calculated using weights that incorporate our own non-response adjustments based on a propensity model. While there are some modest differences, the three sets of estimates are broadly similar. The paper ends with a discussion of survey design features, their effect on the types and level of non-response, and the tradeoffs associated with different design choices.