Time-use data offers unparalleled insights into the daily lives of children, of particular interest for research on child development. Much of this research is based on cross-sectional time-use data, but longitudinal time-use data would be ideal. However, longitudinal time-use data, following the same individuals regularly over time, are rare. The Longitudinal Study of Australian Children (LSAC), uniquely, has collected annual time-use data for two cohorts of children. Specifically, there are three waves of time-use data in the Baby cohort (children 0-1 years in the first wave), and four waves of data in the Kinder cohort (children 4-5 years in the first wave). Primary carers provided time-use data, using a light diary, for the first three waves of data in both cohorts. Children provided the fourth wave of time-use data in the Kinder cohort, again using a light diary.

Methodologically, these data present an opportunity to examine issues of survey non-response, attrition and data quality pertaining to longitudinal time-use data. Exploiting this opportunity, this paper analyses patterns of non-response and attrition, and assesses the representativeness of the data using R-indicators. The paper also considers issues relating to the quality of the data, adopting a holistic approach to considering data quality. This approach advocates the use of all information available in the data about activities, location and co-presence when making assessments about the quality of the data. Analysis of these issues will not only support substantive research using these data, but also provide insights for possible use in any future development of longitudinal time-use surveys internationally.