Abstract:
Social surveys are usually affected by item and unit nonresponse. Since it is unlikely that a sample of respondents is a random sample, social scientists should take the missing data problem into account in their empirical analyses. Typically, survey methodologists try to simplify the work of data users by completing the data, filling the missing variables through imputation. The aim of this paper is to give data users some guidelines on how to assess the effects of imputation on their micro-level analyses. We focus attention on the potential bias caused by imputation in the analysis of income variables and poverty measures. We consider two methods for evaluating the effects of imputation, using the European Community Household Panel as an illustration.