Conferences

The paper proposed focuses on the unit-nonresponse in the IAB Establishment Panel, a comprehensive data set describing the employer side of the labour market in Germany. Every year since 1993 (1996) the IAB Establishment Panel has surveyed the same establishments from all branches and different size categories in western (eastern) Germany. Data are collected in personal interviews with the owners or senior managers of the establishment. Meanwhile more than 15.500 firms participate in the survey. The questionnaire covers a variety of questions on important aspects of the employment. Although great efforts are taken to convince the owner/manager to take part in the survey there are always firms that do not answer the questionnaire. If this unit-nonresponse does not occur at random serious selection problems can arise. Therefore it is important to know about the selection process in order to develop effective methods to reduce nonresponse or to construct useful postsurvey compensation schemes. In this paper we try to find out why some establishments are not willing or able to respond to the questionnaire. We are especially interested in the influence of the interviewer on the survey participation. We find support for the thesis that the mutual trust between interviewer and the firm’s representative is crucial for the decision to participate again in the survey. We also hit upon some evidence that interviewers who carry out a lot of interviews are more successful in
avoiding nonresponse.