Evaluation of weighting methods to integrate a top-up sample with an ongoing longitudinal sample

Abstract

Long-running household-based panels tend to top-up (or refresh) their samples on occasions over the life of the panel. The motivation for adding such samples may range from concerns about overall sample size, lack of population coverage or inadequate samples of small target groups. In 2011, a general top-up sample was added to the Household, Income and Labour Dynamics in Australia (HILDA) Survey, a decade after the original sample was selected. This top-up sample added 2150 responding households to the main sample of 7400 responding households, representing a 29 per cent increase in the overall sample size. These top-up samples can be used to improve the cross-sectional and longitudinal weights. Drawing on the experience of other large household-based panels, this paper evaluates six options for integrating the two HILDA samples. The evaluation considers the variability in the weights and the root mean square error of a range of key estimates.