There is regular non-response in sample surveys which creates bias in estimates of population parameters. The present study develops some techniques to estimate a finite population using second-order raw moments of an auxiliary variable in the presence of non-response in a sample survey. Two estimators of the ratio type and the difference type have been proposed in the two traditional situations of non-response. The bias and mean square errors have been calculated under large sample approximation in both situations: situation-I and situation-II. The performance of the proposed estimators in terms of Percent Relative Efficiency (PRE) is shown in conjunction with other existing estimators discussed in this article using theoretical and empirical studies using real data sets. Concerning the famous Hansen and Hurwitz (1947) method of proposed estimators, the percent relative efficiency is always greater than the percent relative efficiency of the other discussed estimators. The results show that the proposed estimators performed better than the other discussed estimators. The results are interpreted, and appropriate recommendations are made to the survey practitioners.