# calculate effect size from regression table

I am working on power analysis for a project. I want to benchmark effect size from another study. How can I find the effect size (Cohen's d) just by looking at the regression table below? For instance, how would you calculate the effect size from the regression table below:

----------------------------------------------------------------------
(1)
outcome
----------------------------------------------------------------------
treatment            .13***       (.03)
var1                -.07*         (.03)
_cons               -.11**        (.04)
----------------------------------------------------------------------
standard error in parentheses
* p<0.05, ** p<0.01, *** p<0.001
R2                   .24
N                    452
----------------------------------------------------------------------


Do you think this formula is useful? Cohen's d = t-value / √N In this case, the effect size of variable treatment is: (0.13/0.03) / √452 = 0.204. Thus, the effect size overall is low. Do you think this is a satisfactory way to do it?

Alternatively, if I wanted to find the cohen's f or eta squared, what additional information would you need?

Best,

• I don't think you can, without more information. Commented Aug 4, 2023 at 23:03
• What else would you need? Commented Aug 4, 2023 at 23:04
• N, variance (or SD) of the outcome. Commented Aug 4, 2023 at 23:33
• And do you want effect size for the whole model or for a specific predictor estimate? Commented Aug 5, 2023 at 7:28
• Just for the specific predictor estimate Commented Aug 5, 2023 at 14:27