I'm trying to calculate the sample size for a 2(between)x2(within)x2(within) Mixed ANOVA using G*Power 3. As I interested in the interaction effect, I have entered the following values in F tests ANOVA: repeated measures, within-between interaction - A priori: Effect size = 0.25, alpha error = 0.05, Power = 0.8, No. of groups = 2, No. of measurements = 4, Corr = 0.5.
The results showed the total sample size required was 24, which I think was weird because it was smaller than the minimum sample usually recommended (30). Is there anything wrong with the value I entered？
In principal, I would say that you did it correctly, there is nothing wrong with low sample size calculations (e.g. in a two sample t-test with a large effect size of 1.5 you would only require 7 subjects per group, one-sided). On the other hand, small samples tend to overestimate the effect size (e.g. Loken & Gelman, 2017) and may not reperesent a stable estimate for your parameters. Therefore, I would suggest to use larger sample sizes and not to rely solely on the p-value (for which you are doing the power calculation), but to focus on the effect sizes and interpret them, not the significance.
You need to decide what is acceptable statistical power for a given significance test. The rule of thumb of 80% power being reasonable is often bandied about. However, I think it is more sensible to see sample size selection as an optimisation problem, where statistical power is but one consideration, and the cost of collecting additional data is considered
- Badges
- Science topic
- Similar topics
- Cognitive Science
- Thinking