​The aim of this study was to identify the most accurate imputation methods Handling missing data through comparing: Means Imputation, Hot-deck Imputation, and Regression Imputation. In order to attain the purpose of the study a random sample of tenth grad student was chosen, it consisted of (2100) in which they completed their responses on all (30) MCQ in the general Math exam, these were used as a source data, Ten percent (10 %) of the data was eliminated in which it allow the data to meet the condition of Missing Completely at Random (MCAR), The Little Test showed that data is (MCAR). later missing data handled by three imputation methods: Means Imputation, Hot-Deck Imputation, and Regression Imputation. Items and Persons Parameters were estimated by BILOG MG3 using Two Parameter Model. In order to explore the most accurate Methods for Handling Missing Data the following statistical methods were used: Correlation methods, Fisher Z, t Paired Sample Tests, and RMSE. The main results showed that the order of Single imputations methods Handling Missing Data regarding Accuracy is: Regression Imputation, Means Imputation, Hot-deck Imputation.​​