Journal

How does double-blind peer review work, and who sees whose name?

In double-blind peer review the reviewers do not see who wrote the manuscript, and the authors do not see who reviewed it; the editor who decides sees both. The NISO terminology standard approved in 2023 calls this model double anonymized. A randomized trial at the WSDM 2017 conference found that reviewers who saw author names favored papers from well-known authors and top institutions.

An editor from behind covers face-down manuscripts with blank cream slips

What is double-blind peer review?

A review model in which the author's identity is not made visible to the reviewers and the reviewers' identities are not made visible to the author. Both identities remain visible to the editor who makes the decision.

The definition comes from ANSI/NISO Z39.106-2023, Standard Terminology for Peer Review, approved by the American National Standards Institute on June 5, 2023. The standard names the model double anonymized. Under it, the reviewer identity is not made visible to the author, the author identity is not made visible to the reviewer, and the decision-making editor sees both.

The editor's view is the point most authors miss. Double-blind review hides names between authors and reviewers, not from the journal. The editor receives the title page with the names and affiliations, chooses the reviewers and reads their reports with their names attached. That is how an editor checks that a reviewer does not work with the author or compete with him.

The standard treats identity as one of four elements of a review process. The other three are whom the reviewer interacts with, what review information is published and whether comments are allowed after publication. A journal that calls itself double-blind has answered only the first of these questions.

How does it differ from single-blind, triple-blind and open review?

By who sees whose name. In single-anonymized review the reviewer sees the author. In triple-anonymized review even the editor sees neither. When all identities are visible, authors and reviewers know each other.

NISO Z39.106-2023 lists four options for identity transparency. In single anonymized review the reviewer identity is not made visible to the author, while the author's identity is visible to the reviewer; the editor sees both. In double anonymized review neither side sees the other, and the editor sees both. In triple anonymized review neither the reviewer nor the author identity is visible to the author, the reviewer or the decision-making editor. With all identities visible, the reviewer is visible to the author and the author to the reviewer.

The second element, whom the reviewer interacts with, has three options in the standard: the editor, other reviewers and the authors. A reviewer in a double anonymized journal normally writes only to the editor, and the editor passes anonymized comments to the author.

The third element covers publication of the review. The standard lists eleven options for it, among them none, review summaries, review reports with or without the author's or reviewer's opt-in, the submitted manuscript, author and editor communication and the identities of reviewers or editors. The fourth element, post-publication commenting, is either open or on invitation. A journal can be double anonymized during review and still publish the reports afterwards.

Why do journals now say anonymized instead of blind?

The NISO standard replaced blind with anonymized to avoid concerns about ableist terms. COPE changed the wording of its peer reviewer guidelines to anonymised review on 29 February 2024.

The NISO working group explained the choice in the standard itself: the term blind was replaced with anonymized because of concerns about using ableist terms. The procedure behind the words is the same. A double-blind journal and a double-anonymized journal hide the same names from the same people.

The Committee on Publication Ethics followed. Its Ethical Guidelines for Peer Reviewers, version 2 of September 2017, record a language change to anonymised review on 29 February 2024. Older journal pages, author guides and conference calls still use double-blind, so both terms appear in searches and submission systems.

For an author the practical test does not depend on the label. The journal's policy should say who sees the title page, whether reviewers see each other's reports and what is published with the article. A journal that only writes double-blind without these details leaves the author to guess.

What is peer review for, and what does the reviewer assess?

The ICMJE defines peer review as the critical assessment of manuscripts submitted to journals by experts who are usually not part of the editorial staff. Editors use it to decide which manuscripts suit their journals.

The International Committee of Medical Journal Editors sets out the roles of authors, reviewers and editors in its Recommendations, in the section on responsibilities in the submission and peer-review process. Peer review there is the critical assessment of manuscripts submitted to journals by experts who are usually not part of the editorial staff. Under the ICMJE text, editors use peer review to decide which manuscripts suit their journals.

Anonymity changes what the reviewer can rely on. Without names he reads the methods, data and argument as they are written and cannot weigh the reputation of a laboratory or a university. The editor still decides; the reviewer gives an opinion that the editor reads together with the second review.

What must a reviewer keep confidential?

The manuscript and its contents. Under the ICMJE Recommendations reviewers do not discuss the work publicly, do not take the authors' ideas before publication, do not keep the manuscript and destroy copies after submitting the review.

The ICMJE gives the reason: authors can be harmed by premature disclosure of any or all of a manuscript's details. Reviewers therefore keep manuscripts and the information they contain strictly confidential. They must not publicly discuss the authors' work, must not appropriate the authors' ideas before the manuscript is published and must not retain the manuscript for personal use; copies are destroyed after the review is submitted.

Help from a colleague is allowed with disclosure. If a trainee or colleague assisted with the review, the reviewer acknowledges that person in the comments to the editor. The ICMJE also asks reviewers to respond promptly, keep to the agreed deadline and write comments that are constructive, honest and polite.

Confidentiality runs in the other direction too. Under the ICMJE text for journals, editors do not publish or publicize reviewers' comments without the permission of the reviewer and the author. If a journal anonymizes its reviewers, their identity is not revealed to the author or anyone else without the reviewer's express written permission.

When must a reviewer decline a manuscript?

When a relationship or activity could bias the evaluation. The ICMJE asks reviewers to declare such relationships and to recuse themselves where a conflict exists. Journal policies list the typical cases: co-authorship, supervision, the same employer, a financial interest.

Double anonymity does not remove conflicts of interest; it hides them from the author. A reviewer who recognizes a former student's dataset, or a competitor's project, knows more than the author does about why the review may be unfair. For this reason the ICMJE asks reviewers to disclose relationships and activities that might bias their evaluation and to step aside when a conflict exists.

The editor checks conflicts before assigning the reviewers, because the editor is the only person who sees both names. The COPE guidelines describe the shared obligations: journals provide transparent peer review policies, and reviewers conduct reviews in an ethical and accountable manner. Clear communication between them is what COPE ties to consistent, fair and timely review.

Does hiding the author's name change reviewers' decisions?

In one randomized trial it did. At WSDM 2017, reviewers who saw author names gave higher scores to papers from well-known authors and top institutions and bid on about 22 percent fewer papers than reviewers who did not.

The trial was run at the WSDM 2017 conference on web search and data mining. Andrew Tomkins, Min Zhang and William Heavlin split the program committee at random into a single-blind group, which saw author names and institutions, and a double-blind group, which did not. Each paper received two reviewers from each group. The results were published in the Proceedings of the National Academy of Sciences in 2017 under the title Reviewer bias in single- versus double-blind peer review.

In the authors' paper, single-blind reviewers were more likely to review favorably the papers by famous authors, from top universities and from top companies, with odds multipliers of 1.63, 1.58 and 2.10. They bid on about 22 percent fewer papers, and their bids favored top universities and top companies, with odds multipliers of 1.13 and 1.17. For papers with women authors the odds multiplier was 0.78, which was not statistically significant in this study alone; the authors' meta-analysis of seven studies found a significant effect in aggregate.

Anonymity is not perfect, and the authors addressed that. In a 2018 reply in PNAS they wrote that to the extent authors' identities are known, the treatment group comes to resemble the control group, so such leakage would understate the true effect. They also wrote that the result does not guarantee the same effect at other conferences or journals.

How does an author prepare a manuscript for double-blind review?

The names go on a separate title page, and the main text is cleared of anything that identifies the authors: names, affiliations and self-citations. The editor keeps the title page; the reviewers receive only the anonymized text.

The standard defines the result, not the steps: the author identity is not made visible to the reviewer. Each journal turns that into instructions. The editorial policy of IAE Review is one example. The title page is a separate file with the full names of the authors, affiliations, ORCID iD, the contact email of the corresponding author, funding sources and conflict of interest information. The main text is anonymized, with names, affiliations and self-citations removed.

The same title page carries the data a reviewer would otherwise use to identify the author: the grant number, the employer, the co-authors. Keeping it apart is what lets the editor check conflicts of interest while the reviewers work without names.

When the review comes back for revision, the anonymity continues. Under the IAE Review policy the author receives anonymized reviewer comments and submits a revised manuscript and a point-by-point response within 30 days. Further review is normally done by the same reviewers.

How does IAE Review run double-anonymous peer review?

Each research manuscript goes in anonymized form to at least two independent reviewers who assess it separately. The first editorial decision takes up to 7 business days, the decision after review up to 60 calendar days from submission.

IAE Review, the peer-reviewed journal of the International Association of Experts, applies independent review to original research articles, analytical reviews, practical case studies and methodological developments. The Editorial Office selects reviewers by the subject of the manuscript from invited specialists, and checks their competence and conflicts of interest before the assignment. The two reviewers assess the manuscript independently and, whenever practicable, are not told each other's recommendations. Commentaries and book reviews receive editorial review.

A reviewer must decline if he or she is a co-author, supervisor or subordinate of an author, has a personal relationship with an author, works in the same organization, has a direct financial interest in the results or cannot give an impartial assessment for another reason. Reviewers keep the manuscript and all related communications confidential and may not use the manuscript before publication. The Editorial Office does not disclose reviewers' identities without their explicit consent.

The initial screening by the Editor-in-Chief or the handling editor checks scope, formatting, word limits and basic quality within 7 business days. Each manuscript is checked for improper borrowing, with a benchmark similarity level of no more than 15 percent excluding properly formatted quotations. Each reviewer assesses relevance, academic novelty, methodological soundness, quality of presentation and fit with the journal's scope, with a target review period of 21 business days. The decision is one of four: accept without changes, accept subject to minor revisions, return for revision and further review, or reject.

Manuscripts are accepted in Russian or English in DOCX, PDF or LaTeX, with abstracts and 5-8 keywords in both languages. IAE membership is not required and does not affect editorial decisions, and no article-processing charge is imposed. An author may appeal a rejection within 30 days on grounds of procedural irregularity, factual error or a reviewer's conflict of interest.

Sources

Read also

All articles

Journal

Want to publish in IAE Review?

Submit your manuscript in Russian or English as DOCX, PDF or LaTeX through the submission form. Initial decision within 7 business days.

Manuscript

Received

Initial decision within 7 business days

Article typeOriginal research
Reviewers2, double-anonymous
DOI10.68034 via Crossref