3 ms·
Small n is not always a deal breaker, IF the small sample resembles the greater population (which I can't judge in this case). https://www.dailykos.com/stories
by baxtr 7y ago
Small n is not always a deal breaker, IF the small sample resembles the greater population (which I can't judge in this case).
https://www.dailykos.com/stories/2008/11/9/656465/- https://www.dailykos.com/stories/2008/11/9/656465/-
- darawk 7y agoThat and if the effect size is sufficiently large, though I can't imagine n=8 is enough to detect an effect size of the type we're likely looking at here reliably.