In this article, we have derived a new calibration estimator of the population mean in the stratified random sampling under the presence of non response. The proposed calibration estimator has been obtained by making some improvements in the Hansen and Hurwitz (1946) estimator. For this purpose, we have gainfully utilized the information on a single auxiliary variable to obtain a set of new calibrated weights that increase the precision of the estimates. The proposed calibration estimator has been derived using the chi-square type distance function subject to some calibration constraints based on the auxiliary information. The Taylor linearization technique has been used to derive the expression for the variance of the proposed calibration estimator. We have also performed an empirical study based on the hypothetically generated data and real data to study the performance of the proposed calibration estimator. The proposed calibration estimator has been found more efficient than the usual Hansen and Hurwitz (1946) estimator and the recent calibration estimator proposed by Dykes et al. (2015).