This research is to study and compare estimation method for missing data of the independent variables in logistic regression. The methods used to estimate missing data are Mean Imputation (MEAN) , Maximum Likelihood Estimation (MLE) , Pseudo Maximum Likelihood Estimation (PMLE) and The Filling Method (FILL) in the case of two independent variables where only one independent variable is affected by missing values. The comparisons are done under condition of sample size of 40, 70 , 90 ,100 , 200 and 400 ; percentage of missing data of 5%,10% and 15% ; correlation in independent variables of 0 , 0.1 and 0.2 . Initial parameter at β₀,β₁,β₂ = 0.2 and β₀,β₁ = 0.2, β₂ = 1. The criteria of determination is the average difference between the estimates and the true parameter and Average Mahalanobis Distance (AMH). The data for this research is simulated by using the Monte Carlo simulation technique with 1,000 repetitions for each case. The results of this research are as follows : According to the comparison of Bias and AMH from four referred methods , it is found that when sample size is less than 90, MEAN method has a smallest BIAS and AMH. In case sample size is more than 90 , FILL method has smallest BIAS and AMH. The BIAS and AMH decreases when sample size increases because more sample decreases error but BIAS and AMH increases when the proportions of missing data in independents variables increase because increasing missing will decrease sample size. In case of initial parameters of independent variables increases, BIAS and AMH increase. In case of the level of correlation among independent variables increases, BIAS and AMH increases.