Bayesian Analyses for EPP, 10 & 11 December 2026
EPP Symposium 171

Please subscribe as soon as possible to make sure that your stay overnight is guaranteed.

If you are not an EPP member, you will have to pay €500 in advance to cover the costs of the symposium.

The dissertation, generally containing multiple scientific articles, is a major outcome of any PhD project. Many PhD students find writing their first scientific article difficult, but most overcome this difficulty and successfully complete their dissertation. Nonetheless, PhD students (and others) may benefit from developing additional writing skills. We might want to prepare a press release about recent exciting findings. We might be asked to contribute to a popular-science magazine or website. We might consider blogging about our research outcomes and experiences. We might want (or need) to submit a research proposal to a granting agency, in order to fund our next project.

In this symposium we will focus on the art of writing. Participants will learn to write (first) scientific articles more efficiently, and will learn about popular-science writing, blogging, and grant writing.

 

Registration deadline: 12 November 2026

Date
December 10-11, 2026

Location
Landgoed de Horst, Driebergen/Rijsenburg, The Netherlands
www.kapellerput.nl

Organization
Alex Wong (Erasmus University Amsterdam), Angelos Krypotos (KU Leuven)

Program

Day 1 — Foundations and familiar analyses

Time Session
10:30–11:00 Arrival and coffee (venue opens at 10:30)
11:00–12:00 Welcome and why Bayesian?

Brief welcome and introduction to the workshop. What Bayesian inference offers that significance testing cannot. Statistics as updating beliefs in light of data: how priors and the likelihood combine into a posterior.

12:00–13:00 Bayes factors and evidence for the null

Quantifying the relative evidence for competing hypotheses, including evidence that an effect is absent. Interpretation of Bayes factors.

13:00–14:00 Lunch
14:00–15:30 Comparing groups

Bayesian approaches to comparing groups: estimating and testing group differences, follow-up comparisons, and weighing competing models. Introduces a routine repeated in every analysis that follows: visualise the posterior, test how robust conclusions are to the choice of prior, and check the model fit with the data.

15:30–15:45 Break
15:45–16:30 Small sample analyses

What Bayesian analysis can and cannot do when samples are small, similar to typical clinical populations: the role of informative priors and statements about uncertainty.

16:30–17:15 Correlations and recap

Quantifying associations between variables, with the same routine as before. Recap of the day.

Day 2 — Advanced analyses and study design

Time Session
9:30–10:30 Moderation

When effects depend on individual differences.

10:30–12:00 Multilevel modelling

Making full use of trial-by-trial data from experimental paradigms: modelling change over trials whilst accounting for individual differences.

12:00–12:15 Break
12:15–13:00 Working on your own data I

Participants apply the approach to their own data or to the course dataset.

13:00–14:00 Lunch
14:00–15:00 Working on your own data II

Participants apply the approach to their own data or to the course dataset.

15:00–15:30 Reporting and closing

Reporting Bayesian analyses transparently and convincingly for reviewers.