Imputation methods for missing data

Target group and prerequisites:

PhD candidates working (or planning to work) with quantitative data sets. Participants should be familiar with basic concepts in descriptive statistics as, for example, taught in introductory statistics courses.

Description of the seminar:

The aim of the seminar is to give participants an overview of commonly used imputation procedures for missing data. Starting with a short explanation on four general mechanisms of missing data, single and multiple imputation procedures will be presented using practical examples. The advantages and disadvantages of the methods will be discussed along with diagnostic tools to assess the quality of the imputation.

Content:

1. Missing-data mechanisms
2. Methods that discard data
3. Single imputation methods
4. Multiple imputation methods

Learning objectives:

After the seminar, you will be able to do the following:

• understand the different types of missing data processes
• assess the type and potential impact of missing data
• explain the advantages and disadvantages of the approaches available for dealing with missing data

Assignment (mandatory for obtaining 1 ECTS):

To obtain 1 ECTS, participants are required to hand in an assignment based on an analysis of the missing data in their data set. For PhD candidates that do not dispose of an own data set to be analyzed, data sets will be provided. For the assignment, participants are expected to use and discuss at least one of the presented approaches to deal with missing data in their analysis. Participants are free to use their preferred statistical software for the analysis.

References:

Gelman A., Hill J. (2007). Data Analysis using Regression and Multilevel/Hierarchical Models, Chapter 25, Cambridge University Press

Kleinke K., Reinecke J., Salfrán D., Spiess M. (2020). Applied Multiple Imputation: Advantages, Pitfalls, New Developments and Applications in R, Springer

Hair J. F., Black W. C., Babin B. J. and Anderson R. E. (2019), Multivariate Data Analysis, Chapter 2