A dataset with no missing values is said to be a complete dataset. A complete dataset is important during the analysis of microarray data and its classification. The goal of missing value handling algorithm is to address the issue of missing data in the microarray dataset and provide a reliable and accurate estimates of these values. In this work, an enhanced imputation method that combines single and multiple methods is proposed. The proposed method handles the missing values works using the information regarding the missing percentage in the dataset. The proposed method works in two stages. The first stage decides on the imputation method based on its missing rate and the second stage uses the decided method to impute the missing values. Experimental results, using six cancer datasets, proved that the proposed method is highly efficient and was able to improve the classification performance in terms of accuracy and speed.