Single-Blind, Double-Blind, and Open Peer Review: What the Trials Show

Peer review comes in three broad models: single-blind, where reviewers know the authors but not the reverse; double-blind, where both are masked; and open, where identities are disclosed. Randomized trials found that masking author identity did not measurably raise review quality and was often unsuccessful, and that revealing reviewer identity also left quality largely unchanged while making some reviewers decline.

Three models, defined by who knows whom

Peer review models differ mainly in one variable: who is allowed to see whose identity. In single-blind review, the reviewers know who the authors are, but the authors do not know who reviewed them. In double-blind review, both sides are masked, so reviewers judge the work without knowing whose it is. In open review, identities are disclosed, and sometimes the reviews themselves are published alongside the paper.

Each model is an attempt to manage the same tension. Reviewers need enough protection to be candid, and authors need enough fairness that the work is judged on its merits rather than its byline.

Single-blind: the traditional default

Single-blind review is the long standing default in much of medicine and science. Reviewers see the authors and their affiliations, which can help in judging, for example, whether a group had the resources to do what they claim. The concern is that the same visibility can bias the review, so that a famous name or a prestigious institution earns a gentler reading than an unknown one.

That worry is the motivation for masking authors. If reputation is leaking into the judgment, hiding the authors should, in principle, make reviews fairer and better. Whether it actually does is a question that can be tested.

Double-blind: does masking authors help?

This is a testable claim, and it has been tested. A randomized trial across several major medical journals masked author identity for some reviewers and not others, then had editors and authors rate the resulting reviews. Masking did not produce measurably better reviews.

Two findings complicate the picture further. Masking was often unsuccessful, with reviewers correctly guessing the authors a large share of the time, because methods, references, and writing style give a paper away. And the manuscripts hardest to mask were those by well known authors, which are exactly the cases where masking might have mattered most. Double-blind review is reasonable on fairness grounds, but the evidence that it raises quality is weak.

Open review: does naming reviewers help or hurt?

Open review pushes in the opposite direction, revealing rather than hiding. The idea is that a reviewer who signs their report will be more careful, more constructive, and more accountable. Here too there are randomized trials.

Asking reviewers to be identified to authors did not change the measured quality of their reviews, and telling reviewers that signed reports might be posted online did not change quality either. What consistently did change was willingness: identified reviewers were more likely to decline. So openness appears to cost reviewer participation without clearly buying better reviews, though many argue it is still worth it for accountability and trust.

What the evidence adds up to

Put together, the trials deliver a humbling message. The main levers journals reach for, masking authors or unmasking reviewers, do not move review quality much in either direction. Quality seems to depend more on the individual reviewer and the guidance they are given than on the blinding scheme wrapped around them.

That does not make the models pointless. Double-blind review can still be defended as fairer, and open review as more accountable, even if neither sharpens the review itself. The mistake is to expect the blinding model alone to fix the deeper limits of peer review.

Reading a journal's review model as a signal

For a reader, a journal's model tells you what to watch for rather than how good its papers are. Under single-blind review, ask whether prestige might have eased a paper through. Under double-blind, remember that masking is often imperfect. Under open review, you may be able to read the reviews themselves, which is a genuine window into how hard the paper was pushed.

None of these models certifies a result. Peer review of any flavor is a filter, not a guarantee, and knowing which filter was used just helps you judge what it was likely to catch and what it was likely to miss.