2017年7月13日木曜日

Visualization of peer review records

Hi, I'm Yuichi Hayakawa from CSIS (Center for Spatial Information Science), The University of Tokyo. 
I originally posted a Japanese version of this article, but there were some requests to provide it in English as well.
Thanks for giving me the opportunity to provide this article in English, regarding the visualization of peer review records. 

Supposing that many readers of this blog are researchers including students, I would like to ask you one question: 




As a researcher, submitting one's paper is an essential task.
On the flip side, reviewing papers of someone else is also necessary for publishing scientific papers. 

However, the process of peer review is often performed anonymously, and the action itself may not correctly be evaluated no matter how much a researcher takes the time to read and evaluate someone's manuscript drafts.

Even though we recognize that peer review is an essential process for the scientific community, this may lead to the disadvantageous feelings that reviewing is rather time-consuming, cost ineffective. 

To overcome such problems, there is a relatively new service, Publons

Publons, founded on 2012, provides a web service on recording and publishing peer reviews with necessary confidentiality. 
"A mission to speed up science - Publons" 
There are a variety of functions in Publons, including post-publication review. However, for simplicity, here I focus on the normal pre-publication review. 

A researcher can register the basic information of a peer review report that she/he performed, including title, journal name, date of review, and review report itself. An option of making it public or private is available for each review following the policy of the journal and the reviewer. 

Here is my case. 
An overview can be embedded in a web page like this.




In the Profile > Statistics view, the timeline of peer review records is shown (as of July 2017 – just 100!).


Monthly review records

For the old review history, I salvaged past e-mails and files to enter the records – somewhat annoying taking a long time... But finally, I can see my total review records like this. It should be better to record the current review in progress every time it is completed. 

The number of reviews has increased, where their frequency has also risen. 
It had been once every few months, but in recent years, it has become more common to review 3-4 cases per month.
This increase may be either as a result of the personal career development and the general increase in the number of publications.

Impact factor (IF) is often regarded as a standard evaluation index of academic journals, and the number of reviews for journals can be shown like this.


Impact factors of journals reviewed for

In my case, the mode is around IF = 2 to 3. This could be because I most often review for Geomorphology (Elsevier) as a board member.

Also, the trend can be compared with those in the field of earth and planetary science.
Researchers in this field seem to have many reviews conducted in journals with a bit higher IFs than me.

Another funny stat is the "weekly review punchcard" – you can see which day of the week you often submit your review report.


Weekly review punchcard

Though we can read manuscripts at any time, it is necessary to keep the time to summarize review results. The review reports, therefore, tend to be sent more frequently during the latter part of weeks where teaching classes and administrative meetings are relatively few, or on the weekend's evening (the midnight of Sunday – actually Monday).

Without doing anything, peer review is just over when it is returned – this should be, of course, a significant contribution to the scientific community. However, recording reviews would help researchers to motivate their review activities by the visualization of review records. 

By the way, we do not have, in fact, a standard lesson or training methods for reviewing so much. Some opportunities like seminars or workshops are being provided recently, but still not so many.
It is necessary to learn by themselves like an on the job training.
Preparation as a peer reviewer - PEPS Editors Blog (mostly in Japanese, but some references link to webpages in English)
Then, even by just looking at the peer review history of others, visualization of review records would help to know how to proceed with reviews.
It is not competition, of course, so the number or frequency of reviewing is not a big problem.
For the high-quality peer review, we can utilize this kind of system to enhance our motivation and to be evaluated in an appropriate manner. 

By the way, Publons has been very recently (last month) acquired by Clarivate Analytics
Also, for example, a major academic publisher like MDPI is collaborating with Publons on its peer review system.
It seems that the peer review evaluation services of Publons are becoming more popular.

0 件のコメント:

コメントを投稿