[R-meta] Can/should we run follow-up tests on subgroup means

Luke Martinez m@rt|nez|ukerm @end|ng |rom gm@||@com
Wed Feb 9 17:50:46 CET 2022

Dear Wolfgang,

I read (https://www.metafor-project.org/doku.php/tips:comp_two_independent_estimates)
with great interest. I had a quick question.

If we have three independent subgroups, should we first run a
Q_between test to determine if the subgroup means are statistically
different or not, and if they are, then run pairwise z-tests to
determine which pair(s) are statistically different?

Thank you,

More information about the R-sig-meta-analysis mailing list