Good afternoon
I was asked to calculate the effect size for some of data for which I ran a t-test. In short I looked at testis size between two non-reproductive periods (lets call them NR1 and NR2) for seven animals. Theres was no significant difference in testis size. I was told to use
www.campbellcollaboration.org/escalc/html/EffectSizeCalculator-SMD1.php
However, this refers to "control" and "treatment" data sets, of which I have non, besides comparing normal states. The issue with using this calculation is it depends on which period (NR1 or NR2) you use as M1 OR M2. Am I missing something? I would like to use the given site but am unsure which calculation to use then (I have the means/SD/Sample sizes etc with groups having unequal sample sizes)
Hi,
please have a look at the homepage from Daniel Lakens and his publications (see link attached). Especially his paper Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs provides lots of information when to use which effect size (e.g. pooled standard deviation vs standard deviation of the control group). Somewhere on his page you can also find excel sheets to simply calculate all different effect sizes he presents in his paper.
Hope this helps, cheerz
Rainer
https://sites.google.com/site/lakens2/effect-sizes
You can use that calculator. Selection of control and treatment is up to you. Only the sign of Cohen's d will change if you swap NR1 and NR2. It will remain the same in absolute value. You can give the absolute value.
Hi,
please have a look at the homepage from Daniel Lakens and his publications (see link attached). Especially his paper Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs provides lots of information when to use which effect size (e.g. pooled standard deviation vs standard deviation of the control group). Somewhere on his page you can also find excel sheets to simply calculate all different effect sizes he presents in his paper.
Hope this helps, cheerz
Rainer
https://sites.google.com/site/lakens2/effect-sizes
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics