(86) Bayesian Data Analysis – Why Bother?
Date:
Contributors: Huber, S. E. , & Rajh-Weber, H.
Venue: 29th Pre-Conference of the Junior Researchers of EARLI, JURE 2025, Graz, Austria, August 23-24, 2025
Abstract: Have you ever wondered why people continue to interpret p-values as if they tell them how plausible their hypothesis is, even though we are repeatedly taught that this is certainly not their meaning? Or, why researchers seem to spend so much time on computing how probable their data were, if a hypothesis they don’t care about was true, when really, they are interested in what processes might underlie their data? And if methods, that could address such questions, actually exist, why do we apparently hear so little about them in our education? If you answered yes to any of those questions, you are in good company: We continue to ask ourselves the same thing which is why we want to explore Bayesian approaches to data analysis as a complement to conventional frequentist analyses in this workshop. We begin by revisiting the frequentist and Bayesian mindsets using simple, practical examples. Then, we juxtapose strengths and weaknesses of Bayesian methods, as we encountered them in the literature. Finally, we encourage participants to reflect on their own research questions and consider how framing them in either a frequentist or Bayesian way could be beneficial for their intentions.
Find all our workshop materials on OSF (under “Files” in the folder “JURE2025”): Materials