(non-)Bayesian evaluation of informative hypotheses

Semester

Semester 1, 2024-2025

Type of course

Methodological and Practical Courses

Date

October 17, 2024

Location

Utrecht University

---

Duration

1 day

Maximum number of participants

40

ECTS

0.5 EC will be appointed for participation in the complete course

Staff

Rebecca Kuiper (UU)

Content

After shortly highlighting the consequences of the “replication crisis”, participants will be acquainted with i) theory-based, Informative Hypotheses and ii) model selection methods that can evaluate informative hypotheses, namely the Bayes factor and the AIC-type information criterion called GORIC(A). This will be done using concepts and examples; formulas will NOT be a part of this course. 

In the second half, there will be more attention for “hands on” using the R package bain and restriktor (specifically, the goric function). For those who rather not work with R, there will also be lab material for JASP.

Learning goals

Learn to set up, apply, and interpret informative hypothesis evaluation (using Bayesian model selection and GORIC(A)).

Preparation

The course materials can be downloaded from https://github.com/rebeccakuiper/Tutorials/tree/main/KLI%20course. This will be up to date just before the start of the course.

You can prepare by installing - before the start of the course - the software you like to use:

- Install R, RStudio, and the R packages bain and restriktor; 

• R can be downloaded here.
• RStudio can be downloaded here.

If you need to refresh your R skills, do the R hands-on mini course (‘R hands-on mini course.pdf’) available via the link above. 

- If you prefer not to work with R, you can use JASP: https://jasp-stats.org/.