In various surveys, presence of measurement errors has led to misleading results in estimation of various population parameters. This study indicates the effects of measurement errors on estimates of population total and population variance when samples are drawn using systematic sampling technique from a stratified population. A finite population was generated through simulation. The population was then stratified into four strata followed by generation of ten samples in each of them using systematic sampling technique. In each stratum a sample was picked at random. The findings of this work indicated that systematic errors affected the accuracy of the estimates by overestimating both the population total and the population variance. Random errors only added variability to the data but their effect on the estimates of the population total and population variance was not that profound.
Category - Cheruiyot W. KIPKOECH
Technical University of Kenya, Kenya