I am performing a random effects meta-analysis of mean differences and I am calculating effect sizes (hedges g) for each study.

Each study contains 2 groups- 1 with a placebo condition and 1 with a drug condition and we have a continuous psychiatric rating scale measuring the change in symptoms for each group. I have been able to obtain mean change and SD of change for each group for each study.

My query is that in some studies one of the group’s S.D of change is 0 - is this is a problem for calculating the effect size with hedges g?

We are not worried that the S.D is 0 as this does occur with healthy controls using psychiatric scales -especially in a placebo group. The question is can hedges g be used to calculate the effect size where one S.D is 0?


Your Answer

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

Browse other questions tagged or ask your own question.