This paper provides evidence on time-use from the Health and Retirement Study (HRS), a large general-purpose survey that is representative of the U.S. population age 51 and over. The data stand out for its rich set of covariates which are used to present variation in time-use by health and socio-economic status. The HRS interviews about 20,000 persons about a wide array of topics, covering economic status, physical and mental health, family relations and support, labor market status and retirement planning. In 2001, a random sub-sample of 5,000 respondents in 5,000 households was sent a supplemental questionnaire, the Consumption and Activities Mail Survey (CAMS), asking respondents about recall information on time-use and spending. It generated 3,866 responses. In 2003, CAMS wave 2 was mailed to the same 5,000 people. The authors combine these data with information on the same respondents from the core HRS survey. Analyses of basic data quality show very low rates of item non-response overall, with slightly higher rates among groups with lower socio-economic status. They combine the information on time-use obtained for 31 activities into the following categories: economic activity, leisure, housework, sleeping and napping, personal care, exercise, managing finances, time at computer, socializing, helping others and volunteering. They compare the CAMS measures with measures from the American Time-Use Study, and find fairly close correspondence. They present descriptive statistics for these time-use categories by personal characteristics, such as age, sex and health status, illustrating the richness of the HRS data.
