An Empirical Test of a Limited Model for Panel Refusals

Abstract

In this article a limited model for panel refusals is presented and tested. The model contains only respondent characteristics: respondent's background characteristics, interest in the topic of the questionnaire and the respondent's difficulties to perform his or her role adequately. Data from an election panel survey are used to test the model. The results make clear that poorly educated women are more likely to refuse the second interview of an election panel survey. The results also support the basic idea of the model that refusals in a second wave of a panel are more likely to be found among those with low ability to perform the task of a respondent and among those who experienced difficulties in fulfilling their role during the first interview.