Infusing time diary evidence into panel data: an exercise in calibrating time-use estimates for the BHPS

In this paper, we calibrate a set of time use variables for a long-running panel survey (the British Household Panel Survey, BHPS, 1994 2004) from evidence derived from a smaller scale panel survey that collected time use information by both the survey method and the diary method from the same respondents (the Home On-line Study, HoL, 1999 2001). Past research has suggested that the time diary method produces more accurate and reliable measures of time use than the survey approach. The diary approach, however, usually has a low response rate and is not practicable for a large-scale panel study like the BHPS. On the other hand, direct questioning in survey interviews is a relatively flexible approach to collect time use data. We therefore propose a method to combine the strengths of the survey approach and the diary method to produce time use data. The survey part of the HoL study shares the same questionnaire-derived time-use predictor variables with the BHPS. We use regression of the predictors on the time diary data in the HoL study to calibrate time use in the BHPS by multiplying the resulting regression coefficients with the same predictor variables in the BHPS. Then we get a calibrated index of time-use patterns based on BHPS questionnaire items. The calibrated time use variables cover all major categories of daily activities and are available in Wave 4 (1994) to Wave 14 (2004) of the BHPS. They are useful resources for the study of time use practices and the life course.