Effect size Interpretation

[Peter Orangi]

My study is based on evaluation the effectiveness of an intervention by comparing trewated and control groups
I have used the following command to calculate the sample sample size for my study : power twomeans mean1 mean2, sd1() sd2() power(0.8)
and the following command to calculate the statistical power : power twomeans mean1 mean2,SD(1) SD(2) n(g1) n(g2) alpha(0.05)

Now i want to calculate the effect size using the following command: esize twosample outcome, by(subgroup)

Fromthe result i get the Cohen's value is negative as shown:

esize twosample shepempowermentcrp, by(tgroup)

Effect size based on mean comparison

Obs per group:
control group = 2,186
treated group = 2,434
---------------------------------------------------------
Effect size | Estimate [95% conf. interval]
--------------------+------------------------------------
Cohen's d | -.1453443 -.2031667 -.0875062
Hedges's g | -.1453207 -.2031337 -.087492
---------------------------------------------------------

Kindly assist me to understand whether the process i have followed to calculate the sample size, statistical power and effect size are correct.
Also please help me understand on how i can interpret the negative cohen's d value?

[Carlo Lazzaro]

Peter:
take a look at The Stata Blog » Measures of effect size in Stata 13

[Peter Orangi]

Hello Carlo,
I have checked the blog and it clearly explains the concept of effect size and the commands for estimation. This was very helpful. Thanks so much