Hello everyone,
I am conducting a systematic review to see if treatment x is superior in its effect than treatment y. Since most the studies I found did not report interaction effect, I was thinking to calculate the effect size to see if x is superior to y. However, I am not sure if I should calculate effect size within group one and compare it to the effect siye withing group two. or should I calculate pre-post effect size between groups?
I look forward to hearing from you
Best,
Ghazal Naser
You should calculate pre-post effect size between groups
Hi Ghazal,
are you planning in performing a conventional pairwise meta-analysis? Than, I believe you will be using the standardized mean difference (SMD), which can be interpreted with Cohen´s d classification of effect sizes.
I am not sure if I understood correctly. Are you considering pooling either post-scores or change scores (from baseline)? The latter is sometimes difficult as not all authors report the change score. Also, (post-hoc) calculating the SD for the change score is difficult and sometimes not possible.
Maybe you could elaborate on your methods.
Greetings,
AJ
- Badges
- Science topic