3 ms·
This shows some general weaknesses of observation studies: - Participants couldn't accurately estimate the amount of alcohol they drank - People change their
by hubraumhugo 5y ago
This shows some general weaknesses of observation studies:
- Participants couldn't accurately estimate the amount of alcohol they drank
- People change their behavior in response to their awareness of being observed(Hawthorne effect)
- Moderate drinking correlates with a healthier lifestyle
- acchow 5y agoI find observational studies only useful in suggesting what to investigate next. It’s not helpful to build theories or anything resembling science on.
- MandieD 5y agoYeah, I've long thought that the "light drinking is healthy" effect was more like "people who regularly drink alcohol but stop at one are probably restrained in their other consumption, while there are plenty of non-drinkers who have terrible diets and have trouble putting down the fork."
- robwwilliams 5y agoIt is a Mendelian randomization based study of observational data. This is much more that JUST observational. See Pearl’s Book of Why on how causality can be tested by MR.