71 undergrads and 43 grads, one university, nothing normalized for anything except hours on facebook and self-reported GPA.
It's hard to even pull that much since it appears that this is only a poster presentation.
I'd love to see a real study of 1000's of students from colleges and universities around the world. Even better, to also consider tools other than Facebook, but I'd settle for just the one tool.
I don't doubt that there is some potential for the results to be accurate but there's way too little information and way too many variables that haven't been taken into consideration. And the story is being picked up by the wire services as "proof" that social tools are harmful.
I'm curious how this affects the sort of safe harbour-ish thinking that says you're legally safe if you don't know what's happening and if anyone can use you for good purposes. But hanging out a kill-switch means you expect to review and approve (or, potentially, remove) apps which means Google would suddenly be liable for anything written with Android???
PL/I -- "the fatal disease" -- belongs more to the problem set than to the solution set. -- Edsger W. Dijkstra, SIGPLAN Notices, Volume 17, Number 5