In this project, we study the development of the LISS panel sample over years. The LISS panel is a large-scale online panel established in 2007 on the basis of a random probability sample from the Dutch population. As a result of recruitment nonresponse and attrition over time, the LISS panel might have some biases in its composition. In an earlier paper, we have estimated the R-indicator for the LISS panel initial sample, using population register auxiliary information, to study the representativeness of the panel for the target population. In the current project we follow a different approach, estimating R-indicators across panel waves by means of the first wave variables. The R-indicator in this case does not indicate population representativeness but panel composition compared to the net sample in the first wave. The research question we aimed to answer in this paper is how systematic the attrition in the LISS panel is with respect to the core research variables.
01.05.2016 - 31.12.2017 / SHARE - Research
Sample composition development in an online panel
Presentations and Contributions