Hello,
I have problems with the data from one study in a meta-analysis. When I manually enter the mean, standard deviation and total data for my two groups into the RevMan software, the automatic weight calculation is wrong. We can easily assume this error given the small sample size and the large size of the green square in the attached figure. How can I correct the error? Thank you very much for your advice
Hi Laura,
I'm not familiar with the RevMan, but i have done similar work with the software Jamovi. Where the mean difference method can be conducted with the MAJOR module. It's all available as open science.
Maybe trying a different software can solve your problems, i hope i helped in some way!
This is not a software issue and the weight is not wrong. That study is weighted so highly because, when the between study variability is negligible as it is here, the weight is determined completely by the variance of the study. Because the standard deviation is so low, that study is extremely precise and therefore has a high weight. This is likely an issue with the data extraction, the measurement scale or incorrect imputation of the sd. It looks mostly like they are on different scales, in which case you would use the standardised mean difference (SMD) not the mean difference.
To be entirely sure, I have reproduced this in different software, which gives you the same results.
- Badges
- Science topic
- Similar topics
- Biological Science
- Anatomy
- Tissue