Winter Course: (Bayesian) Informative Hypothesis Evaluation

About

This e-learning course introduces informative hypotheses: a powerful alternative to traditional null and alternative hypotheses. With a strong emphasis on application rather than formulas, the course offers a hands-on learning experience using R and JASP. You will gain practical skills and conceptual understanding to apply these techniques effectively in your own research.

Since Cohen’s (1994) influential paper “The Earth Is Round, p μ₂ > μ₃

H₂: μ₁ – μ₂ > μ₂ – μ₃

In recent years, interest has also grown in alternatives to null-hypothesis significance testing.

This course introduces two such approaches for evaluating informative hypotheses:

Model selection with the AIC-type criterion GORIC(A), using GORIC(A) weights;
Bayesian model selection, using Bayes factors and posterior model probabilities (PMPs).
Both methods quantify the relative support for competing hypotheses by balancing model fit and complexity. For example, if the GORIC(A) weight ratio or Bayes factor comparing H₁ to H₂ equals 5, the data provide five times more support for H₁ than for H₂—suggesting H₁ is the best current description of the population.

In this e-learning course, you will learn what GORIC(A) weights and Bayes factors (and PMPs) are, how to obtain them in R and/or JASP, and how to interpret them.

The course is designed for PhD students, junior lecturers, and researchers who want to deepen their understanding of this methodology and apply it to their own data.

The course emphasizes concepts and examples over formulas, offering a clear and applied learning experience. It is structured as follows:

Day 1 (January 15): Online kick-off meeting to get started.
Self-paced learning: Course material available for seven days.
Day 7 (January 22): Online Q&A and lab session

Course leader

Dr. Rebecca Kuiper

Fee info