This paper addresses the challenges of estimation of population mean for small or no sample size in the presence of nonresponse and presents a calibration estimator that produces reliable estimates under stratified random sampling from a class of synthetic estimators using calibration approach. Examining the alternative estimator under three distributional assumptions, namely, normal, gamma, and exponential distributions through a simulation analysis with average absolute relative bias, average coefficient of variation, and average mean squared error as evaluation criteria, the results show that it has a consistent estimates of the mean with less bias and greater gain in efficiency. Further validation through the coefficient of variation also shows that the estimator exhibits a more preferred coefficient of variations suitable for small area estimation.
Keywords: calibration, distance measure, nonresponse, small area estimation, synthetic estimators.
