In conducting a meta-analysis, sometimes, some studies have reported geometric mean, while others have reported arithmetic mean. One solution for pooling these studies is calculating the ratio of the mean; I wanted to know if the ratio of the mean calculated as the following: mean intervention/mean control? and what should we do with standard deviation? Can we then pool the ratio of arithmetic mean and the ratio of geometric means?
Ratio of means is mean intervention/mean control.
It can be done with geometric or arithmetic means.
while it is possible to approximate arithmetic from geometric and vice versa. this is not recommended. nor is mixing in a meta analysis.
There is other correspondence on this on RG and the attached/references are important for methods
Article The ratio of means method as an alternative to mean differen...
Article Ratio of geometric means to analyze continuous outcomes in m...
Dear James,
Thank you so much for your reply and the suggested papers. I read the papers;
is my conclusion right "We can pool RoM and RoGM only when RoGM has been calculated with the Taylor series method"?
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis