6 ms·
You're down voted but I agree. 200 participants split into A/B for yes/no drugs, 8 scenarios and 3 different tests makes the sample incredibly small.. would lik
by itcrowd 9y ago
You're down voted but I agree. 200 participants split into A/B for yes/no drugs, 8 scenarios and 3 different tests makes the sample incredibly small.. would like to hear statistical arguments opposing my PoV
- hammock 9y agoPretty standard for this type of research
- tbabb 9y ago"Everybody does it" is not really a statement about its validity.
- ricardobeat 9y agoStraight from the paper: > we determined sample size based on previous research which indicated that a sample size of about 40 participants per cell provides sufficient power to detect a behavioral effect of acetaminophen (Durso et al., 2015). For Experiment 2, a power-analysis based on a power criterion of (1−β) = 0.80 and effect sizes obtained in Experiment 1 indicated that a mean cell size of n = 54 was sufficient to replicate significant findings