Almost no literature exists on the effects of the topic of the questionnaires on panel participation. In this paper I will explore how the addition of “tailor-made questions” at the end of existing questionnaires will affect the evaluation of the questionnaire. The experiment will be embedded in the Dutch CentERdata LISS panel; a panel of 5.000 households that complete surveys every month.

The panel members were asked if they could name three topics they are most interested in. These topics are classified into about 40 categories. The experiment we will implement includes five versions. In the first version respondents will receive a set of questions at the end of the questionnaire about a topic they prefer. The second version will be equal to version one, except that the questions are about a topic they did not mention as one of their preferred topics. In the third version, the panel members will be presented 3 to 5 topics, selected from the list of 40 topics. These 3 topics include a topic the panel members mentioned as one they are interested in. Version four will be the same but there will not be any topics included that the panel members referred to as a topic they are interested in. The fifth version will be the control version; these panel members will evaluate the questionnaire without the tailored question part. In this experiment we will explore the effect of extra questions (about a topic the respondent is interested in) and the influence of the option to choose a topic from a list.