Missing data in sampling surveys

Abstract

The main aim of this dissertation is to study the problem of missing data in surveys. This study will lead us to analyze other issues related to nonresponse. In fact there are other aims in this dissertation. One of them is the discussion about the use of the characteristics of the population and sample, mainly demographic issues. These will be used for the implementation of imputation models. Imputation will be divided in two types; imputation to cater for respondents who refused to answer more than one question in the questionnaire, but not all of them and imputation to cater for those persons who refused to answer all the questions in the questionnaire. These analyses will be done with respect to two different kinds of surveys but three surveys for the analysis. Our main interests relate to be the classification of political tendencies of the nonrespondents. Another problem studied related to the estimation of income of nonrespondents. Finally we are going to compare and analyze the results obtained in this dissertation and also, further recommendations for further study.