#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

# How to calculate Effect Size in Bayesian Repeated Measures ANOVA

Hi everyone,

I'm currently using JASP for my master thesis and am trying to figure out how to calculate the size of effect in a Bayesian Repeated Measures ANOVA.

I'd like to report the results like this: BF10 = x, R^2 = y

I read the following in another Paper that used Bayesian Repeated Measures ANOVA: "Effect sizes were computed as the increase in R^2 when adding the factor to the model."

How exactly am I supposed to do this and which values do I need?

Would be thankful for some input/ideas! 🙏

• Hi,

Your queries concern two crucial aspects of Bayesian Repeated Measures ANOVA: the Bayes Factor (BF10) and the effect size (R^2).

1. Bayes Factor (BF10): When performing a Bayesian Repeated Measures ANOVA in JASP, it automatically calculates the Bayes Factor. This value shows the strength of the evidence supporting the alternative hypothesis over the null hypothesis. A larger BF10 shows more robust evidence for the alternative hypothesis.

2. Effect Size (R^2): JASP does not directly calculate the R^2 value for Bayesian ANOVA. In a frequentist approach, one would compare the R^2 values from two models–one with the factor of interest and one without. The increase in R^2 illustrates the variance explained by the added factor. To compute Bayesian R^2, you could consider utilising R packages, such as 'BayesFactor' or 'rstanarm'. These offer functionalities to calculate Bayesian R^2 values.

I hope this helps!