Bayesian ANOVA

Mateus

ANOVA.PNG

Hi everyone,

I am trying to understand the results of a Bayesian ANOVA.

I performed a cluster analysis (in another software) to cluster my sample according to training variables (e.g., distance run, accelerations, etc.). Then, I performed a Bayesian ANOVA in Jasp to quantify the probability of the clusters created being related to well-being variables (e.g., sleep time). I considered the clusters as a "fixed factor" and each new variable as a "dependent variable".

How can I interpret these results? My data provide strong support for the for no relationship between training performance and sleep time?

I attached the figure.

Thanks in advance,

Mateus

EJ

Hi Mateus,

Yes, good evidence for the null model and against the "cluster" factor

E.J.

Mateus

Hi E.J.,

Thank you for taking the time to write to give me some feedback.

One last question. Although, the good evidence for the null model, should I describe the posterior distributions? (e.g., Cluster 3 revealed higher posterior distribution on sleep time compared with the other clusters).

I look forward to hearing from you.

Thanks in advance,

Mateus

EJ

Hi Mateus,

Opinions differ. You could show them and indicate explicitly that these obtain under the assumption that there is an effect (an assumption that the data undercut).

EJ

Mateus

Hi EJ

Thank you for your time.

Mateus