Room: Kerkzaal
Day 2 — Methodological and Bayesian Advances
|
09:00 – 09:30
|
Walk-in & coffee
|
|
09:30 – 10:45
|
Session 4 — Bayesian Design and Power for Multilevel Studies
This session focuses on Bayesian approaches to study design and statistical power in multilevel research, including simulation-based design strategies and methods for planning complex hierarchical studies.
- How do I start planning my multisite project? A tutorial on multilevel simulation-based power and sensitivity analysis
Alicia Franco-Martínez (Universidad Autónoma de Madrid, Spain)
- Bayesian Power Analysis for Multilevel Models – the BayesSSD package
Ulrich Lösener (Utrecht University, The Netherlands)
- Sample Size Determination for trials with the Bayes factor
Camilla Barragan Ibanez (Utrecht University, The Netherlands)
- Bayesian sequential designs in studies with multilevel data
dr. Mirjam Moerbeek (Utrecht University, The Netherlands)
|
|
10:45 – 11:15
|
Coffee break |
|
11:15 – 12:55
|
Session 5 — Computational and Bayesian Methods for Multilevel Models
This session presents recent methodological developments in multilevel modeling, including advances in causal inference, variable selection, Bayesian model specification, and computational approaches for hierarchical models.
- Variable selection via knockoffs for clustered data
Prof. dr. Silvia Bacci (University of Florence, Italy)
- Post-selection inference in Linear Mixed Models
Anna Nikolei (University of Kassel, Germany)
- Eliminating rank deficiency in Bayesian models with the ZeroSumNormal distribution
Adrian Seyboldt (PyMC-labs, The Netherlands)
- Dynamic Extended Generalized Many-Facet Rasch Model for Rater Cognitive Bias
Dr. Giuseppe Mignemi (Kore University, Italy)
-
A Multilevel Psychometric Framework for Experimental Tasks: The Generalized Hierarchical Factor Model
Ricardo Rey-Sáez (Universidad Autónoma de Madrid, Spain)
|
|
12:55 – 13:45
|
Lunch |
|
13:45 – 15:00
|
Session 6 — Missing Data, Causal Inference, and Evidence Synthesis in Multilevel Models
This session addresses challenges related to missing data, model estimation, and evidence synthesis in multilevel studies, including multiple imputation approaches and meta-analytic methods.
- Multiple Imputation of Missing Data in Longitudinal Designs: A Comparison of Different Strategies
Mark Lustig (University of Hamburg, Germany)
- Multiple imputation of multilevel data with single-level models: A fully conditional specification approach using adjusted group means
Prof. Dr. Simon Grund (University of Hamburg, Germany)
- Mixed-effects location-scale models in meta-analysis
Dr. Wolfgang Viechtbauer (Maastricht University, The Netherlands)
- Meta-Analytic Pooling of Intraclass Correlation Coefficient Estimates
Dr. Bethany Hamilton Bhat (Maastricht University, the Netherlands)
|
|
15:00– 15:30
|
Drinks & snack break |
|
15:30– 15:40
|
PhD Award Ceremony |
|
15:40 – 16:40
|
Closing Keynote Lecture – Prof. dr. Paul Bürkner (TU Dortmund University)
Amortized Bayesian inference for multilevel models |
|
16:40 – 16:45
|
Closing remarks |
