Inhaltspezifische Aktionen

Early Career Researcher Workshop: Introduction to Bayesian Inference: A Hands-on Workshop for Early-career Scientists

Bayesian methods are increasingly common across the empirical sciences, but getting started can be daunting. This two-day workshop led by Dr. Julius Pfadt and Henrik Godman (University of Amsterdam) introduces Bayesian inference for researchers who work quantitatively and would like to add Bayesian tools to their repertoire. No prior experience with Bayesian statistics is required.

  • Early Career Researcher Workshop: Introduction to Bayesian Inference: A Hands-on Workshop for Early-career Scientists
  • 2026-09-17T10:00:00+02:00
  • 2026-09-17T16:00:00+02:00
  • Bayesian methods are increasingly common across the empirical sciences, but getting started can be daunting. This two-day workshop led by Dr. Julius Pfadt and Henrik Godman (University of Amsterdam) introduces Bayesian inference for researchers who work quantitatively and would like to add Bayesian tools to their repertoire. No prior experience with Bayesian statistics is required.
Wann

17.09.2026 von 10:00 bis 16:00 (Europe/Berlin / UTC200)

Alle Termine

17.09.2026 von 10:00 bis 16:00
18.09.2026 von 10:00 bis 16:00

Wo

Ze-S3, Senckenbergstraße 03

Termin zum Kalender hinzufügen

iCal

 

Überblick

  • Donnerstag, 17. September 2026, 10:00 - 16:00 Uhr
  • Freitag, 18. September 2026, 10:00 - 16:00 Uhr
  • Ze-S3, Senckenbergstraße 03, Gießen
  • Studierende, Wissenschaftler/innen auf jedem Level
  • Englisch

Beschreibung des Workshops

 

Both workshop days combine conceptual input with hands-on work in JASP and in R. JASP is a free, open-source statistics program with a graphical interface.

Day 1 builds the foundations: what a posterior distribution is, where priors come from and how much they matter, and how Bayesian inferences differ from frequentist ones. Participants fit the basic analyses themselves and see the benefits of the Bayesian framework in action. Peer exchange gives space to relate the material to participants' own research.

Day 2 moves to more involved topics: Model comparison/averaging using a regression example, and how to check and report results. Participants work through example datasets from different empirical fields. The day closes with an introduction to R-based Bayesian tools for models beyond the graphical interface of JASP.

Learning objectives — Participants will learn:

  1. What a posterior distribution represents and how prior, likelihood, and data combine to produce it
  2. How Bayesian inference differs from frequentist inference
  3. How to choose priors and assess how sensitive their conclusions are to that choice
  4. To fit and interpret standard Bayesian analyses in JASP
  5. How to check whether a model has converged and whether it describes the data adequately
  6. Report a Bayesian analysis in a form suitable for publication
  7. Know which tools to turn to for models beyond the scope of the workshop

Recommended prior knowledge

Familiarity with standard quantitative methods (t-tests, correlation, regression) from applied use. No prior experience with Bayesian statistics required. Part of Day 2 uses R, prior R experience is not strictly needed and all code is provided.

Software

JASP (free and open source) for the hands-on work on both days. On day 2 we will be using R in RStudio. Instructions for installing the required R-packages will be sent to registered participants in advance of the workshop.

Preparation

·       Install JASP before the workshop: https://jasp-stats.org/download

·       Install R and RStudio

·       Bring a laptop

Instructors
 
Dr. Julius Pfadt is a post-doctoral researcher in the lab of Eric-Jan Wagenmakers at the University of Amsterdam and a former DFG Walter Benjamin Fellow. His research focuses on Bayesian statistical modeling, with particular expertise in psychometrics, reliability estimation, and structural equation modeling. He develops open-source tools for JASP, including modules for Bayes factors, factor analysis, reliability, and structural equation modeling, as well as the R package Bayesrel. He is also a co-founder of JASP Services BV and is committed to open science and accessible methodology.
 
Henrik Godmann is a PhD candidate in the lab of Eric-Jan Wagenmakers at the University of Amsterdam. His research focuses on robust Bayesian inference, interrupted and nonlinear time series, state-space models, and statistical methods in the behavioral sciences. He develops Bayesian state-space methodology for intensive longitudinal data and interrupted time series, with a particular interest in quality control and accessible statistical methods. He is also a co-founder of JASP Services BV and is involved in the development of open-source software for applied researchers.

 

Anmeldung

https://eveeno.com/bayesian-inference