The main objective of the presentation is to answer the question significant for social survey research why some people participate in a scientific panel survey while others refuse to participate. The so

‐called unit‐nonresponse, i.e. the systematic denial of a person to participate in a scientific survey, might be an important problem in social research, since it can be a crucial source of selective sampling. Meta‐analysis of existing literature on unitnonresponse provide empirical evidence that the empirical analysis of the reasons of unitnonresponse are not driven by testing the theories explaining the individuals’ refuse in participating at a survey but it is based on so‐called ‘variable sociology’. That WAPOR Annual Conference 2009 4 ‐participants have been used to deduce the main reasons for individuals’ unit‐nonresponse in a speculative way. Social decision‐making mechanisms of the respondent’s decisions with some explanation power working behind these distributions thus remain undetected. To answer the central research question, the causes for the participation or non‐participation in surveys have to be uncovered. ‐nonresponse as a result of an individual decision of the respondents. In traditional panel surveys, there is a lack of essential variables explaining the individuals’ participation behaviour; therefore it is necessary to conduct these information in a special research design. ‐mail addresses in order to send the results to them. However, these email addresses has been used to ask the former respondents to participate in an online survey on drug use twice. The actual participation is explained by the previously collected theoretical constructs. The collected data provide following results: At the beginning of a panel survey, considerations of costs (such as fear of data misuse) as well as the factor of the respondent’s currently available time have influences the respondents participation decision. Respondents will take part at the second wave of a panel survey more likely if only a few questions were stressful in the initial wave and, moreover, if the survey topic is of interest for them. These results support the theory of rational action in order to explain unit‐nonresponse. As part of this study on the participation behavior in social science surveys, an experiment was carried out to explore for a mailed questionnaire empirically, whether non‐monetary incentives increases the cooperation of the respondents in an classroom survey and reduces their fear of costs. Therefore, two experiments were carried out. ‐monetary incentives (one 50g Toblerone) had no influence on the cooperation of the respondents, a promised non‐monetary incentive (lottery) leads to the opposite effect. It appeared to be successful for the respondents cooperation, although could not reduce the respondents fear of costs.

means that the distributions of demographic variables of participants and non

One model of participation behavior in social research – the theory of rational action (based on the theory of subjective expected utility) by Esser (1974, 1984, 1986, 1990) – explain the unit

Using an experimental design, the theoretically relevant expectations, evaluations and attitudes toward scientific surveys has been collected. The sample contains about 300 students at the University of Berne (classroom survey; random selection of courses). At the end of the questionnaire, students were asked to provide their e

Students of the first experimental group received the questionnaire combined with a bar of chocolate (here: one 50g Toblerone). Students of the second experimental group were offered the chance to participate in a lottery (here: 2 x 2000g Toblerone). While the use of prepaid non