I have data from 9 separate studies and would like to estimate the average mean and standard deviation across them. I was planning to use the -meta- commands to do so and wanted to check this is OK or if there is a more efficient way to do this:
Code:
cls clear all set obs 9 input means sd n -9.8 5.9 36 -5.2 3.9 13 -6.0 8.6 103 -8.4 8.4 102 -10.7 8.2 103 -11.2 8.5 103 -11.6 8.7 102 -13.4 8.2 102 -15.8 8.5 103 g se = sd/sqrt(n) meta set mean se meta summarize * Estimate mean and SD across studies ci mean mean sd [aweight=_meta_weight]
