I love when people say "x is heavily correlated with y" as if the strength of an observed correlation is the key to a causality Things can be very 'heavily' correlated and have no causal connection at all!
-
-
Incidentally, for
#epitwitter spurious correlations dot com is a great place to get examples to use in slides about caution in causal inference (you can even make your own correlation graphs!)Show this thread -
(I mean, we can all make our own correlation graphs, but there's a quick and easy tool to correlate random datasets and produce a nice graph that even prioritizes by strength of association it's very cool)
Show this thread
End of conversation
New conversation -
-
-
In the pocket of Big Cheese I see
-
I used to be paid by Big Pasteur but since raw milk has fallen out of fashion I had to earn the income back somehow

- Show replies
New conversation -
-
-
Totally valid observation. However, I feel it should be said that everything that is causative is also correlated. So, correlation is not the end of the journey to proof, but possibly the beginning.
-
That's true, but in this world things that are causally related are often not correlated on simple observation because everything is so wonderfully complex
End of conversation
New conversation -
-
-
That's excellent. I quite liked this one, too:https://twitter.com/Datasaurs/status/967927394811219968 …
-
That's gold
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.