Help with statistical power?

mgzx

New Member
#1
Hi, I'm doing a study looking at the ability of a test to detect a change in bacterial numbers pre- versus post-treatment [i.e. not treatment effect, but pilot study looking at efficacy of the test used]. I've done a chi-square test and got a significant result (p<0.05). However given that I have a small sample size of 10 patients, is my study considered underpowered? And how do i calculate the ideal sample size for such a study?
 

rogojel

TS Contributor
#2
hi,
power should not be an issue here aa you have found an effect. (Power being the probability of detecting an effect).Given the small sample I would worry more about the reproducibility -i.e. if you repeated the test with a new sample of 10 would you see the same effect size?
regards
 

mgzx

New Member
#3
hi,
power should not be an issue here aa you have found an effect. (Power being the probability of detecting an effect).Given the small sample I would worry more about the reproducibility -i.e. if you repeated the test with a new sample of 10 would you see the same effect size?
regards
Ahh I see, that makes a lot of sense, thanks a lot! :):)