this is true in many fields - in a study I saw, one field the blinded reviewers could correctly identify 40% but in some fields much higher - Christina Pikas

Christina, did that study have anything to say about the perceived quality of the final published manuscript? That is, did the author change the way she amended the manuscript based on her knowing to whom she was responding? - Marie from FreshFeed

I don't remember that twist. Sometimes authors suggest reviewers, in that case, they'd probably be more likely to recognize the comments. I've seen from editors where they get obnoxious responses from one side or the other... There are a lot of questions about the value of peer review and quality of papers, but I haven't seen someone test if the authors knew the reviewers how that changed things - Christina Pikas

Maybe for something as narrow as described here--but when I've done peer review for ITAL and thought I recognized the author, I've been consistently wrong. And I write any comments in a studiously anonymized style. So, in general, I don't buy this. - Walt Crawford

So regardless of whether you're been right or wrong about who the author or reviewer is, does imagining that you know who it is change your comments or revisions? I think it would have to. - Marie from FreshFeed

hmm...also thinking about your mention of ITAL, Walt. A journal like this, with a broader scope, may not provoke the same issues. - Marie

Marie: Honestly? No, it didn't. And if I'm a good reviewer, it can't: A friend of mine can write a crappy paper (that's certainly happened!); an enemy of mine can write a superb paper (I don't know of many enemies, but that could certainly happen). I did note "something as narrow as describe here"--but even most library niches have at least 30-40 people, enough that I'd doubt identifiability. - Walt Crawford

Walt, I think the way you approach manuscript review is The Way It Should Be Done. - Marie