Nonresponse Patterns and Bias in the American Time Use Survey

• Activity Number: 491
• Type: Topic Contributed
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): John Dixon*+
• Companies: Bureau of Labor Statistics
• Keywords: survey ; nonresponse ; measurement error ; total survey error ; time use

Abstract

The American Time Use Survey (ATUS) is 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 collects a wealth of demographic information about respondents, this design provides information about ATUS nonrespondents. Measurement error, due to either forgetting an activity or mistakes estimating the time or duration, may also contribute to bias in the ATUS survey. This paper focuses on nonresponse bias and measurement error. A propensity score model is used to examine differences in time-use patterns between those who are likely to respond and those who are reluctant, and to assess the extent of nonresponse bias. The two processes (forgetting and duration recall) will be explored by looking at the large number of zeros in many time-use categories, with the zeros serving as indicators of possible forgetting, and unusual durations conditional on remembering indicating possible recall error.