repeated measure ANOVA - missing data & multiple-groups
Hi!
I have the following dataset:
12 patients x 4 sessions each
12 controls x 2 sessions each
However, it happens that some patients have incomplete data. For example, some might have just 2 sessions or 1 session out of 4. I am having trouble finding the best statistical test for this particular case. Can someone suggest the most appropriate statistical test to apply in order to find differences between groups and sessions, considering that I have missing data? I thought repeated measures ANOVA would be suitable, however, I am concerned with the missing data.
Thanks! I will appreciate any input or suggestion!
ARF
Comments
Hi ARF,
If you have many missing data you might want to look into missing data imputation. This is something that we still have to do in JASP, but R has several packages for this. I vaguely recall that data missing at random are not a problem for our Bayesian repeated measures ANOVA, but I could be wrong. Let me know if you want to explore the Bayesian route and then I can look into this more deeply.
Cheers,
E.J.
Hi E.J,
Thanks for your suggestions! Yes, I would like to try Bayesian repeated measures ANOVA.
Thanks!
ARF
Though advanced, I believe a Linear Mixed Model would be an appropriate alternative to repeated measures ANOVA, since the former doesn't require that patient have complete data.
R
The BayesFactor models are in fact linear mixed models. Let me get confirmation from Richard, also on the missing data aspect.
Ok, perfect! Thanks!
ARF
Yes, so Richard confirms that it would just be an unbalanced design in that case, which the Bayesian model handles.
Ok, thanks! I will try it!