Non response is a common problem in a survey process. Therefore, it is necessary to find a way out of handling non response whenever arises. The current study proposed difference-cum exponential ratio-type estimator for estimation of general parameters using auxiliary information which is defined in two situations of non response. A conventional estimator t*(a,b)  is used to define population constants including population mean, standard deviation, coefficient of variation and mean square. The expression of bias and mean square error of the proposed estimators are obtained up to the first order of approximation for situation I and situation II. To compare the efficiency of the proposed estimators over the existing ones, an empirical study is carried out using real and simulated data sets. Both the theoretical and empirical study shows that the proposed class of estimators outperformed other existing estimators.