The present research is an effort to study the importance of measurement error in detail. The main aim of the study is to estimate the finite population mean in the presence of measurement error under non-response and randomized response while using simple and stratified random sampling schemes. We developed a generalized class of estimators by using dual use of auxiliary information in simple and stratified random sampling. Efficiency comparison is done theoretically and numerically. For numerical comparison both simulation study and real life data sets are used in this study. To gain the practical insight of measurement error, data is also collected and analyzed. In Chapters 2 and 3, the problem of measurement error and non-response in simple and stratified random sampling is studied. In Chapters 4 and 5, we discussed the estimation of finite population mean for sensitive variable in the presence of measurement error under simple and stratified random sampling. In Chapters 6 and 7, we studied those practical situations in which non-response and measurement error both occurs while estimating the finite population mean for sensitive variable, under simple and stratified random sampling, respectively. In Chapter 8, two stage cluster sampling for estimating the population mean in the presence of measurement error and non-response is studied.