In 2009, the Panel Study of Income Dynamics (PSID), an ongoing national panel study of a sample of U.S. households from 1968 to present, included a 24-hour time diary supplement, which was administered to 400 couples with at least one spouse age 60 and older. Spouses were each interviewed twice (one weekday and one weekend day) and interviews were coordinated so that husbands and wives interviews yielded same-day diaries. Like the American Time Use Study (ATUS), the PSID diary interview was administered by telephone with a computer-assisted interview. However, unlike ATUS, the PSID CAI had several unique features designed to minimize potentially biasing, directive probing by interviewers. Permission was obtained from 716 out of a possible 757 respondents (95%) to record part of the diary interview. All of the recordings cover the first four diary activities, making it possible to characterize the interactions between interviewer and respondent associated with the recall of particular activities and affect during those activities as well as temporal estimation processes. This proposal seeks funding to develop a coding scheme and then code the diaries. Analytic files will then be created and linked to public use files that contain measures of data quality. Data analysis will be carried out and results presented at a professional meeting. Findings from this project will help inform how to enhance quality of future time diary efforts, particularly around issues of how much to script time diary CAI applications, approaches to handling verbalizations of uncertainty by respondents, and identification of new probes to provide interviewers or to include in web-based (interactive, self-administered) applications.
Funding:
Princeton University
Funding Period:
09/01/2010 to 08/31/2012