Missing Data Imputation

Abstract

Missing data in clinical research data is often a real problem. As an example, a 35 patient data file of 3 variables consists of 3 × 35 = 105 values if the data are complete. With only 5 values missing (1 value missing per patient) 5 patients will not have complete data, and are rather useless for the analysis. This is not 5 % but 15 % of this small study population of 35 patients. An analysis of the remaining 85 % patients is likely not to be powerful to demonstrate the effects we wished to assess. This illustrates the necessity of data imputation. Imputed data are not real data, but constructed values that should increase the sensitivity of testing. Regression imputation is more sensitive than mean and hot deck imputation, but it often overstates sensitivity. Probably, the best method for data imputation is multiple imputations (4), because this method works as a device for representing missing data uncertainty. However, a pocket calculator is unable to perform the analysis, and a statistical software package like SPSS statistical software is required.