Howdy, Stranger!

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

Supported by

Is it possible to run Bayesian ANOVA with an unbalanced design?

Hi,

I am running a serial position experiment, where participants remember varying numbers (1-4) of items, and we probe their memory for items in different positions of the study set. This means the design isn't fully balanced, because for example you can probe positions 1-4 when 4 items were presented, but can only probe positions 1-2 when only two items were presented. So, our design looks like this:

Set size Position probed
2 1
2 2
3 1
3 2
3 3
4 1
4 2
4 3
4 4

When I create two factors in the repeated measures bayesian anova window, it tries to set up a fully crossed design like this:

Set size Position probed
2 1
2 2
2 3
2 4
3 1
3 2
3 3
3 4
4 1
4 2
4 3
4 4

This leaves me with empty cells in the design which I can't possibly fill. Is it possibly to get around this?

Comments

  • Last I checked, all Bayesian analyses (ANOVA and regression) in JASP (which are done with BayesFactor) are actually liner-mixed models, so unbalanced designs should not be a problem.

  • In principle this should not be a problem, but I suspect that situations such as these require the data to be in long format. We are working to make that happen but we're not there yet. Good point this; you could post this issue on our GitHub page so it won't be forgotten.
    E.J.

Sign In or Register to comment.