help > Linear Mixed Effects Models for longitudinal data analysis
Showing 1-1 of 1 posts
Display:
Results per page:
Jul 12, 2019  07:07 AM | Lisa MP
Linear Mixed Effects Models for longitudinal data analysis
Dear Conn experts,

I would first like to thank you for your tool, which has been very helpful to analyze the resting state fmri data I am working on.

I would like to ask you a little question about second-level test design on longitudinal data.

The data I work on has two groups of subjects, and for the moment two timepoints (soon three). I tried the following contrasts:
Subjects effects : cov1 ... covn GroupA groupB
Between subjects contrasts : 0...0 1 -1
with
Conditions : Timepoint0, Timepoint1
Between conditions contrasts : -1 1

I want to test the difference of longitudinal change in connectivity between the groups A and B. When I enter this contrast, does Conn look at the difference in connectivity (Timepoint1-Timepoint0) and use this difference in a multiple regression model, or does it use something like a Linear mixed-effects model?

Thank you very much for your help,

Best,

Lisa