Hi all,

I have two (hopefully) simple questions. I want to compare the studies in my meta-analysis on some characteristic of those studies. For example, I want to test if there is a difference between studies that use a social stimulus (i.e., pictures of faces) and compare the meta-analytic results against studies that use more basic stimuli (i.e., just simple geometric shapes). These are all studies with a single group and whole-brain data.

Is this a good fit a linear meta-regression? All of the examples I've found on the forum are using some continuous variable.

If I can use linear meta-regression to test this, is there anything I should know about setting up the table? Could I code the studies as, for example, 1 if using social stimuli and 0 for basic?

Thanks!
dan

Hi Dan

If I've understood what you want to do it seems the compare groups linear model function may be best - https://www.sdmproject.com/manual/?show=...

As you said, code one group in the table as 0 and the other as 1. If you called this column in the table "group", for example, chose group in the analysis window to define the two sets of studies and it will give you a mean map for each group as well as the contrast between them.

Hope that helps,
Jo