Missing Data and Multiple Imputation

This chapter reviews the problems of missing data, randomness of missing data, and multiple imputation. Missing data have the potential to bias results; thus, whenever possible, a case that is missing a limited amount of data should be included in the dataset. Multiple imputation is a procedure for estimating the values of missing data, enabling cases with some missing data to be included in the dataset. Using real data from the high school and beyond (HSB) dataset, this chapter shows the steps for assessing patterns of missing data to explore if the data are missing at random. Next, the process of restructuring and imputing the data is presented. Step-by-step guidance on how to conduct these processes is included, along with screenshots and outputs, and how to write about the results in APA format is included.