5 Reasons you should probably just say no to a peer review request

You probably agree that the peer review system is broken. Several ideas have been proposed to fix the system. Some journals tried double blind peer reviewing and others experimented with open peer review.

However, I think these proposals are like putting a Bandaid over a cancerous wound because they don’t really address the root cause of the problem.

Scientists often pride themselves as a highly objective contingent. Yet, research on single blind peer reviews show women and researchers from low and middle income countries are often viewed unfavorably. Thus, double blind reviews might help, a little.

But double blind peer review has its problems. Sometimes, a reviewer needs to look at the previous work by an author, especially when understanding the context of the research is important. Open peer review might help by taking transparency to the limit.

But there’s a problem with open peer review. If scientists can succumb to bias, nothing can stop them from falling into revenge and vindictiveness. The science world is very small; I’m a postdoc, giving a negative review to a senior researcher has dire consequences to my career.

The answer to peer review challenges is probably going back to basics. And different polls agree – to fix peer review, peer reviewers need formal training. It is often assumed graduate advisors teach their students how how to conduct peer review – most don’t. Hence, publishers have developed formal programs to teach peer review.

I think formal training is good and should be encouraged. Training is like removing the cancerous cells. It addresses the heart of the problem. And I recommend anyone who cares about science to take an online class on peer review. You will thank me letter.

I came across a paper in Environmental Research that had wrong chemical structure of compound, poor experimental design, and numerous logical errors. Reading the paper, it was obvious the original manuscript lacked a solid peer review.

In a more serious case, a prominent researcher submitted a manuscript in his son’s name. I noticed this because, he cited himself about 15 times out of 40. Importantly, the son had zero research experience in the subject area but the father did. Anyone who actively researched on the topic of the paper should have noticed the red flags.

I’m certain that if some of the reviewers who reviewed these two manuscripts had rejected the request and let competent researchers do the job, these manuscripts would have been rejected.

Reviewers should learn to say, NO. Period. I think this should be the easiest decision considering that as a reviewer you probably spend 5 unpaid hours preparing a review for a journal that will probably charge you $50 to read that same article.

But if you need additional reasons to text the next request to review, I have five for you. Say no:

  1. When you’re busy and are certain you won’t be able to read the manuscript at least twice.
  2. When you think the paper is boring after reading only the abstract.
  3. When you have doubtful knowledge on the subject matter.
  4. When you are friends or mortal enemies with one of the authors.
  5. When the results of the study affect your financial interests in any way.

Published by

Edmond Sanganyado

Edmond Sanganyado (PhD, University of California) is a postdoc in China. He is interested in the effect of organic pollutants in aquatic environments. His work has been featured at Publons, The Good Men Project, and University of California Riverside's Gradsuccess blog.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.