Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

High correlations with immense statistical significance don't imply any one path of causation, but they do imply causation is involved. That doesn't mean A causes B, it may mean D causes A and B, etc. But mere coincidence isn't plausible when your statistical base number in the millions and the effect is large (as this was.) We agree that correlation and causation aren't the same word or concept - I never said anything different, nor would.

I'm proposing a possible explanation (as you have also done!), not claiming proof; you seem to be confusing a claim of evidence (which may indeed be false or misleading, but remains evidence) with a claim of proof. These are different things, esp in Science post-Popper, although conflating the two is very popular, even by scientists, today! I haven't said other explanations are impossible. That's always so. But you might want to argue for your view by finding other supporting evidence.

For example, pro your view, a much better argument on your side is that there was no possibility in those days of controlling for fetal alcohol syndrome - it hadn't been discovered yet. In fact Vernon's studies of Aboriginal IQs also preceded that discovery and FAS turned out to explain most of the lower IQs effect he measured amongst Native populations.

I've added an unecessary "!" here just 'cause you did.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: