Why do most journals not use a double blind peer review evaluation for manuscripts?

Peter Gorsuch explains the different models for peer review, and why they are used.

Go to the profile of Peter Gorsuch
Apr 13, 2019

Want to get access to this course?

To get started, Register for your free Nature Masterclasses account. Signed up previously? Sign in here


Go to the profile of Zhou Li
Zhou Li almost 2 years ago

sometimes, knowing who did the experiments is important :-)

Go to the profile of Sergio Lopez
Sergio Lopez over 1 year ago

It would be better to disclose absolutely everything: both the identities of the authors, which are anyway obvious most of the time, and the identities of the reviewers. I have heard stories of reviewers rejecting a paper after a protracted period of time simply because they wanted to gain enough time to reproduce the experiments and publish the results before the original authors. It seems that in some cases this sort of nasty behaviour actually allows people to steal other researchers' work. Hence, if the identities of all of the people involved are disclosed, reviewers will be held accountable for what they do and for what they write. I suspect that most journals do not disclose the identities of the reviewers because less people will be willing to review a paper. Yet, I think that journals should offer the reviewers some sort of incentive to compensate for the disclosure of their identities during the review process  (i.e., offering them free online courses or something like that). It will be expensive, but I think that the scientific community deserves it. 

Go to the profile of Victor S Garcia Rea
Victor S Garcia Rea about 1 year ago

Agree with you, especially in the compensation part. The peer review process should provide benefits also to the reviewer. A review takes highly valuable time, and even when the reviewer is collaborating with the improvement of science (I hope), another reward could be beneficial, for example the courses that you mention.   

Go to the profile of Lynne Bush
Lynne Bush over 1 year ago

I think double-blind is better, given inherent bias regarding gender that seems to creep into even the best scientific minds. It would be interesting if you did an experiment the with sending out the same paper for review with two fictitious teams of researchers assigned as authors--one male-dominated and one female-dominated. 

Go to the profile of Victor S Garcia Rea
Victor S Garcia Rea about 1 year ago

Nice experiment, another nice one would be trying with different nationalities and Institutions.

Go to the profile of Franziska Eller
Franziska Eller 5 months ago

Go to the profile of Franziska Eller
Franziska Eller 5 months ago

This has already been done, at least similarly, with an application within scientific faculty. The "male version" of the application (which was equal to the female's application, except for the gender of the applicant) was favored and judged better by both faculty males and females. So yes, scientist are absolutely likely to subconsciously exhibit bias against females, therefore also against manuscripts authored by females. Double-blind or transparent peer review are the only ways to minimize that bias, imho.

See the publication on the experiment her:

PNAS October 9, 2012 109 (41) 16474-16479; https://doi.org/10.1073/pnas.1211286109 

Go to the profile of MOMANYI Kelvin Nyariaro
MOMANYI Kelvin Nyariaro about 1 month ago

I would prefer more of double-blindling with continuous training offered to the reviews. Training could be in terms of feedback to them.