Estimating effect size in power calculation

[Chris Labos]

Hello all,

The answer to this question is probably no, but I had a quick question after doing a quick power calculation today.

power onemean 514, sd(117) power(0.9) n(300)

When trying to compute the effect size given a certain number of patients, the automated output is the standardized mean difference as below.

Estimated effect size and target mean:

        delta =    0.1878
        ma =  535.9671

Now I realize it's pretty easy to figure the difference in means here. It's just 536-514 = 22. But I almost made a pretty big booboo today before I caught myself. I don't see any options to output the difference rather than the mean difference, but figured I would ask.

Thanks in advance,

Best,

Chris

[Larissa Zhu]

Hi, I have a follow-up question. I am now using vce2way for regression and need to get Cohens' d to check for the effect size. However, I tried "estat size" and an error popped up by saying "estat esize only works with vce(ols) ". Could you please assist me with that?

Thank you very much.

[Dirk Enzmann]

To answer #1: You specified the null (hypothesis) mean at 514. To see the mean difference in the output specify the null (hypothesis) mean at 0.
Remark to #2: Note that the term "Cohen's d" is ambiguous, see http://dx.doi.org/10.13140/2.1.1578.2725 . To answer #2 others have to jump in.