Methods used in reduction of errors arising from nonresponse

Various types of survey errors, especially nonresponse errors, may seriouslydeteriorate data quality. Nonresponse has more one reason to worry about its harmfuleffects on the survey estimates. So, nonresponse and methods dealing withnonresponse have increasingly become a standard part of survey sampling. Anonresponse occurs in a survey when, for any reason, a selected unit does notrespond. The usual methods of estimation in the presence of nonresponse give biasedresults. Because of this special estimation techniques are required to deal with theproblem. Imputation and reweighting are two standard methods provided by theliterature for treating nonresponse. In the recent years, scientist became increased toconcern with the calibration approach to reweigthing method in the presence ofnonresponse. The calibration approach generates the final weights which are as closeas possible to specified design weights, while respecting known auxiliary populationtotals or unbiased estimates of these totals. This calibration procedure requires theformulation of a suitable auxiliary vector, through a selection from a possible largerset of auxiliary variables.In this study standard methods for the reduction of bias and errors arising fromnonresponse are explained. The calibration approach is examined as theoretically andsimulation is performed by macro generated in C++ programming language to studyhow alternative specifications of the auxiliary vector affect the quality of estimatorsderived by the calibration technique. In the application, the calibration estimators andquality measures such as relative bias, variance are computed and interpreted for thepopulation total.Keywords: Nonresponse, nonresponse errors, calibration, auxiliary information,nonresponse adjustment, nonresponse bias, weighting