Workshop on Statistical Methods for Missing Data Analysis
28 maj 2025 09:30 – 15:00 Örebro University, room to be announced soon
The presence of missing data is a common issue in statistical data analysis that can result in biased parameter estimates and reduced statistical power. The amount of bias and reduction in power depends on the percentage of missing values in the dataset, the missing data mechanism, and the missing data handling method that should take into account correlates of incomplete variables (i.e., auxiliary variables).
Welcome to a Workshop on Statistical Methods for Missing Data Analysis!
The workshop is arranged by the LEADER research environment and will be given by Dr. Takuya Yanagida from University of Vienna.
Dr. Takuya Yanagida is a senior scientist at the Faculty of Psychology, University of Vienna. He gave more than 50 workshops on various methodological topics, including Bayesian statistics, latent variable modeling, multilevel modeling, longitudinal data analysis, mixture modeling, and missing data handling at universities worldwide. Dr. Yanagida has been an associate editor of the European Journal of Developmental Psychology for more than eight years and has been a member of the editorial board of the International Journal of Behavioral Development since 2024.
Content
The goal of the workshop is to provide an overview of three modern methods for handling missing data: Full information maximum likelihood (FIML), Bayesian estimation, and multiple imputation. The main focus of the workshop is on (1) descriptive statistics for missing data and identifying auxiliary variables in R and (2) FIML estimation and incorporating auxiliary variables in the analysis model using Mplus.
The following contents are covered in the workshop:
- Missing Data Mechanism and Auxiliary Variables
- Missing data mechanism (MCAR, MAR, and MNAR)
- Auxiliary variables
- Full Information Maximum Likelihood (FIML) Method
- Factored regression specification for dichotomous predictors
- Saturated correlates model
- Introduction to Bayesian Approach for Missing Data Handling and Multiple Imputation
Prerequisite
Basic knowledge of regression analysis is required, practical experience with the statistics programs R and Mplus is an advantage, but not required.
Literature
Enders, C. K. (2022). Applied missing data analysis (2nd ed.). The Guilford Press.
Hayes, T., & Enders, C. K. (2023). Maximum likelihood and multiple imputation missing data handling: How they work, and how to make them work in practice. In H. Cooper, M. N. Coutanche, L. M. McMullen, A. T. Panter, D. Rindskopf, & K. J. Sher (Eds.), APA handbook of research methods in psychology: Data analysis and research publication (2nd ed., pp. 27–51). American Psychological Association. https://doi.org/10.1037/0000320-002
Software
Mplus Version 8.5 or later, R version 4.3.0 or later, RStudio version 2024.04.0 or later, and the latest versions of the R package misty is required to work through the examples and exercises.
Other info
Fika will be provided. Lunch is at ones own expense.
Questions can be directed to Sevgi Bayram-Özdemir.
Registration
Please register to the workshop before 11 April.