I am doing a meta-analysis and one of the papers report d instead of SD. I do also have sample size and mean for each group, but I don't have the SD for each group.


A bit of simple algebra should suffice here.

Let $X_1$ and $X_2$ be the means which you say you have, $d$ be Cohen's $c$ and $s$ be the pooled standard deviation.

$$ d = \frac{X_1 - X_2}{s} $$

Multiply both sides by $s$

$$ s * d = X_1 - X_2 $$

Now divide both sides by $d$

$$ s = \frac{X_1 - X_2}{d} $$

which is the desired result.

Note that this is more complex if what they are reporting is not in fact Cohen's $d$ but the bias corrected version usually known as Hedges' $g$.


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.