Before starting a case-control study on pharmacogenomics, we have to calculate the sample size. Some researchers advised to calculate the sample size by using the odds ratio of a published article. There may be a situation where no previous odds ratio is available. There is also a post study method for detection of sensitivity.
You can estimate the sample size that will give you enough statistical power, from the number needed to treat (NNT) that you would expect to make sense when using your drug. I guess that kind of information you can either find from similar drugs, or just simply making your own educated guess. When your sample size makes your study statistically significant, but when your NNT is so high that no physician in their right mind will prescribe your course of treatment, it means your sample size was simply too high to begin with.
I agree with Alexei. Also, you can use this software free available on line
http://www.gpower.hhu.de/
There are many statistical formula to calculate sample size according to the population you included in the study:
Sample size from finite population can be estimate by :
n= N / ( 1+ N * e^2 )
and from infinite population by :
n= ( Z^2 * S^2) / E^2
where : Z ( 1.96 for 0.05 and 2.58 for 0.01 )
S = standard deviation from previous studies or pilot study
E = significant level
Also, you can use software to calculate sample size ( SPSS , Minitab , Gpower )
I hope that be useful for you
All of the solutions are fine. Khalid Hassan proposes a theoretical solution. Jose Avila-Peltroche gives a quick practical means to give you an answer.
In order to give you an advice in this issue it is better to know first the hypothesis of your study.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size