Nonresponse Bias for the Relationships Between Activities in the American Time Use Survey

John S. Dixon


The American Time Use Survey (ATUS) is the first continuous, Federally-funded survey designed to measure how people spend their time. The ATUS sample is drawn from households completing their final month of interviews for the Current Population Survey (CPS). Because the CPS records contain a wealth of demographic information about respondents, this design enables us to look directly at nonresponse without having to rely on techniques such as data matching or the use of reluctant respondents to model nonrespondents. Our paper focuses on nonresponse rates and nonresponse bias. First, we describe nonresponse rates by demographic characteristics, and then we use logistic analysis to examine correlates of nonresponse, including demographic and interviewer characteristics. A propensity score model is utilized to examine differences in time-use patterns and to assess the extent of nonresponse bias.