I am working with Rev manager 5.3, doing a meta-analysis on 8 studies that each compare 2 psychological treatments and found no difference between them.
This is a tricky question. You need to do meta-analysis on difference in means.
Usually studies have reported these data:
Case group: mean and SD before, mean and SD after, difference in means and SD of difference in means
Control group: mean and SD before, mean and SD after, difference in means and SD of difference in means
you need to use difference in means and SD of difference in means for each group. If studies have not provide the difference in means it can be calculated very easily by Difference in means=mean after-mean before
The problem is for SD of the before after difference. If it is not reported, you should use the Cochran method in the link below:
https://handbook-5-1.cochrane.org/chapter_16/16_1_3_2_imputing_standard_deviations_for_changes_from_baseline.htm
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis