Interpretting main effect * covariate interactions in Bayesian RM-ANOVA.

Dion · September 2016

Hi, I am a new JASP user (by the way, really enjoying it so far, especially on my more archaic hardware). I have a question regarding the JASP output for Bayesian RM-ANOVAs.

My design involves two within-subject variables (A and and a covariates (c) within a classic RM-ANOVA the default model specification leads to the following output:
(Main effects and interactions with covariates): A, Ac, B, Bc, AB, AB*c

The same specification within the Bayesian variant leads to A + c and A + B + A* c rather than simply A + Ac? This implies the main effect for both A and covariate c, right? I have to explicitly ask for the interaction Ac? And since it is a covariate would I be looking to compare the models A to A + Ac or to A + c + Ac?

Thank you!

EJ · September 2016

Hi Dion,
Can you give us some more detail regarding the hypothesis? And maybe you can add a .jasp file or some output table? (if you aren't ready to share your results with the world just create fake labels and/or use half of the data set).
Cheers,
E.J.

Dion · September 2016

Hey E.J.,

Thanks for the quick response. I've attached a JASP file (sorry for the unclear labels, I'm borrowing the data). The first analysis in the JASP file is a standard RM-ANOVA, the second is the Bayesian. The hypothesis would be that the data supports the interaction Information * NVS.

Thank you,

EJ · October 2016

Hi Dion,
I can't see the file -- maybe I'm looking in the wrong spot. You can Email it to me directly, or put it on the OSF and send the link.
Cheers,
E.J.

Howdy, Stranger!

Categories

Interpretting main effect * covariate interactions in Bayesian RM-ANOVA.

Comments

Howdy, Stranger!

Quick Links

Categories

Interpretting main effect * covariate interactions in Bayesian RM-ANOVA.

Comments