Read on Twitter
Psych Stats/Methods people: Tomorrow, hypothesis testing strikes back in my undergrad research methods class. Wanted to share some resources that I've found really helpful to talk about Bayesian hypothesis testing. /begin Thread.
In preparation, students are reading this helpful blog post by Richard Morey ( @richarddmorey) and they'll be using Morey's BayesFactor R package in class, https://richarddmorey.github.io/BayesFactor/ .
They're also watching a couple of Crash Course ( @TheCrashCourse) Statistics videos hosted by Adriene Hill ( @adrienehill). and
During class, we'll analysis data from our project this semester, first using null hypothesis significance testing (independent samples t test and effect size), then using Bayesian hypothesis testing (BF).
I'm especially excited to show them how we can update our belief about the probability that the alternative hypothesis is true going from original published study to our first direct replication to our second direct replication. To that end, we'll use...
Tom Faulkenberry's ( @tomfaulkenberry) terrific Shiny app. The user inputs a t stat, Ns, & the prior probability that the null hypothesis is true. It calculates Bayes Factors and posterior probabilities for both the null and alternative hypotheses. https://tomfaulkenberry.shinyapps.io/bayesFactorCalc/
Tom Faulkenberry also published this useful lecture video that walks students through the logic of Bayes Factors and how to use the Shiny app.
Looking forward to class tomorrow. Guess I better finishing prepping. Hope some of this is helpful for others. I also welcome suggestions for other resources I could use, either immediately or in the future. /end Thread
P.S. Forgot to link in @richarddmorey's helpful blog post. Here 'tis: http://bayesfactor.blogspot.com/2014/02/the-bayesfactor-package-this-blog-is.html
I'll add resources suggested by others to the tail end of this thread so others can easily find them. First up, JASP stats, suggested by Phil https://twitter.com/Jawsy_P/status/1328004483637411841?s=20
Second up, a recent column written by Tom Faulkenberry to introduce folks to Bayesian hypothesis testing. https://twitter.com/tomfaulkenberry/status/1328014444094513154?s=20
