Some good suggestions and tips in the post and if I may, I'd like to add a tip of my own: scan the CV/resume for spelling, grammar, formatting mistakes.
I know it's not a very "scientific" and objective criterion, but it has never failed me :-)
Do... you invite people who've made those mistakes in to interview, at least? Else I'm not sure how you could have even a fairly unscientific sense of whether applying this criterion has failed you or not.
Definitely, and to be honest I’ve seen a few exceptions that confirm my unscientific rule. I try not to judge by the look, but in a CV for an engineering role I like a nice format without errors. Same goes for code reviews. I prefer clean & commented code over dirty/sloppy hacks.
Ah, cool. I definitely would expect some significant correlation there, but wouldn't be so sure without checking—looks like you're Doing It Right (well, as right as one can with hiring practices, short of running an actual study), sorry for implying you might not have been.
I know it's not a very "scientific" and objective criterion, but it has never failed me :-)