Youth Nonresponse in Panel Surveys: investigating the impact of life events

Abstract

Nonresponse in surveys is particularly problematic among young people in both cross-sectional and panel studies. This article investigates the factors driving lower participation rates among young people in longitu- dinal surveys. We study whether nonresponse can be explained by young people experiencing more life events associated with disengage- ment from panel surveys. Using data from the Understanding Society panel in the United Kingdom, we employ a discrete-time multinomial logistic hazard model to study nonresponse across panel waves. Consistent with previous research, our analysis identifies lower educa- tion, unemployment, immigrant background, and residential circumstan- ces as key predictors of nonresponse. Furthermore, we demonstrate that changes in employment status and (expectations of) residential reloca- tion significantly contribute to predicting attrition among young partici- pants, with age remaining a significant factor. Living with parents also plays an important role, as it is associated with a lower risk of non- contact.

Publication
Journal of Survey Statistics and Methodology

Related