This workshop provides an introduction to handling missing data in statistical analyses via multiple imputation in Stata. We begin with a discussion of missing data mechanisms and how model results might be affected if we use complete case analysis. We then discuss how multiple imputation reduces bias and produces more accurate estimates of uncertainty. Along the way, we provide practical advice for conducting multiple imputation in Stata, emphasizing the importance of understanding assumptions, selecting appropriate models, and carefully evaluating the imputed results.
