Howdy, Stranger!

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

Supported by

Multiple linear regression across different groups

edited July 2017 in JASP & BayesFactor

My dataset has 17 subjects divided into 2 different groups (2 different types of one disease). For each group I have a repeted measure of the variable of interest - mean connectivity. I would like to perform a multiple linear regression analysis across these groups with 7 covariates in total. Is there a way to perform this in JASP that after I get the results I could associate them with a certain group? Basically, I'm interested in evaluate if the differences in the values of PCC_MC, SMA_MC and IPS_MC are correlated with one of the several covariates. But if I have a significant result, for example, cov1 will I be able to identify which of the two groups are associated to that result?

For another dataset I would like to do the same (multiple linear regression) but at this time without the repeated measures. In this case I only have the variable of interest - mean FA and the same covariates. I want to perform one analysis between subjects and discriminate the results for each group.

Thanks a lot for your attention,


  • Hi AnaF,

    We haven't implemented MANCOVA yet, but otherwise you could deal with your mean connectivity DV one variable at a time and run three ANCOVAs. This is only a suggestion after a brief consideration of the problem, there may be other methods that are better.

    With relatively few subjects and many covariates, you might run into trouble.


Sign In or Register to comment.