33 people per group is enough for statistical significance with moderate effect size, which they found here. If you don't think it's legit, get a counter-study

33 people per group is enough for statistical significance with moderate effect size, which they found here. If you don't think it's legit, get a counter-study published. Or go into the math and show how they used the wrong statistical test. Peer-review exists, and the people doing it are a lot smarter than "um sample size lol."