I am about to start a randomized controlled study consisted of 3 groups. I want to know how I can use Gpower software to calculate the appropriate sample size for this study?
Hello Hisham,
The answer depends on:
1. What outcome(s) will be assessed (and how they are quantified); continuous scores would be analyzed differently from categorical (e.g., success/failure) variables.
2. How small a difference is worth detecting, if in fact it exists? (This is the target effect size.)
3. How much risk of being wrong are you willing to have? This concerns both the risk of a type I error (false rejection of a true null hypothesis, quantified as the alpha level), and the type II error risk (failing to reject a false null hypothesis, quantified as "beta," the complement of which is statistical power. e.g., power = 1 - beta).
As a simple example: In G*Power, a fixed-effects, one-way anova ("F tests" family), with a desired power of .90 or better, alpha of .05, target effect size (Cohen's f) of 0.25, and 3 groups would require 207 or more cases in all (69 per group).
If that seems too daunting, I am certain that your institution has a number of folks who can walk you through the process of selecting a suitable sample size for your research aims.
Good luck with your work.
See this paper
Article How to Calculate Sample Size in Randomized Controlled Trial?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size