Daniel Engber is sick of uppity web commenters pointing out that correlation isn’t causation:
And thus a deeper correlation was revealed, a link more telling than any that the Missouri team had shown. I mean the affinity between the online commenter and his favorite phrase — the statistical cliché that closes threads and ends debates, the freshman platitude turned final shutdown. “Repeat after me,” a poster types into his window, and then he sighs, and then he types out his sigh, s-i-g-h, into the comment for good measure. Does he have to write it on the blackboard? Correlation does not imply causation. Your hype is busted. Your study debunked. End of conversation. Thank you and good night. [Slate]
Engber leads with a terrible example. The Missouri team in question found that depressed college students sent more email, shared more files by P2P, and jumped between web applications more often than their happier peers.
Some commenters tried to discredit the study by noting that correlation doesn’t necessarily imply causation. The thing is, the researchers didn’t set out to elucidate the causes or effects of depression. They wanted to see if unobtrusive internet tracking could flag users at risk for depression.
Maybe depression causes farting around on the internet, or vice versa. Or maybe some third factor, like unemployment, causes both. For the authors’ purposes, it doesn’t matter. If they can predict depression from internet habits, they might be able to use that association to identify depressed students. Their odds of success depend more on the size and strength of the association than the reasons behind it.
In this case, the correlation/causation critiques were beside the point because the researchers weren’t making a causal claim. So, some self-important people misapplied some generally sound advice. Welcome to the internet.
I’ve read Engber’s piece three times and I’m still not sure what his beef is. He notes that sometimes correlations are clues to causal relationships. No kidding. Nobody wants a moratorium on investigating correlations. Nor should we reflexively dismiss correlational studies. Documenting an association can be the first step towards understanding a phenomenon.
We should be skeptical of causal claims based solely on correlation, not least because bad faith actors are constantly bombarding us with bad causal arguments based on correlations.
It’s only a matter of time before the RIAA trumpets the Missouri study as proof that filesharing causes depression. When that day comes, the correlation/causation point will be right on the money.
SPECIAL DEAL: Subscribe to our award-winning print magazine, a publication Bernie Sanders calls "unapologetically on the side of social and economic justice," for just $1 an issue! That means you'll get 10 issues a year for $9.95.
