6 ms·
You're describing an observational study, which has significant scientific problems compared to a randomized study. Sure, you can end up with a lot of non-drin
by greglindahl 11y ago
You're describing an observational study, which has significant scientific problems compared to a randomized study.
Sure, you can end up with a lot of non-drinkers, but you will have to attempt to correct for the fact that they probably won't be similar to drinkers.
A randomized study starts off with 2 very similar populations. You still have to worry about dropouts skewing the results, but at least you started off on the right foot.