1
$\begingroup$

How to calculate pre and post treatment Hedge's g effect size to use in a meta-analysis for studies with no control group? Can you do it with just pre and post treatment means and standard deviations and the sample size or do you need a correlation statistic?

$\endgroup$
1
$\begingroup$

You would need the correlation as well. If some studies report the correlation you could impute the correlation from them to the other studies. If there is no evidence about correlation available to you then trying the calculation for a range of plausible values would be possible. You would have to report the results from all of them obviously, picking the one which gave the result you liked best would not be a good idea.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.