Do your review instructions ask if data+software are available?

It looks like PLOS Biology doesn’t ask reviewers to help uphold their data availability policies…. and I’m sure they aren’t the only journal missing this step.

I just send this email to PLOS Biology.  When you review a paper, check the material you are sent to see if you are asked to assess appropriate availability of materials, and if not (or not with sufficient emphasis) please make your voice heard.  You are welcome to use my dashed-off email as a template if it helps, needless to say.

Hi PLOS Biology,

I’m reviewing a paper for you now.  I’ve just realized that your email to reviewers contains several important prompt questions, but no prompts asking us whether data+software have been made appropriate publicly available, as per PLOS guidelines or community norms, whichever are stricter.

Sections 5 and 6 in your reviewer guidelines don’t cover this either… actually it doesn’t seem covered by your reviewer guidelines at all.

Your author instructions say:  “All appropriate datasets, images, and information should be deposited in public resources”… but there does not appear to be any reviewer check?

Seems a pretty big lost opportunity: reviewers are very well placed to make recommendations about what data should be made available.  The “detailed protocols” mentioned in your reviewer guidelines are unlikely to suggest datasets or software to most people.


a big fan of data
Update:  a few days later, PLOS Biology responded as such (and gave me permission to post their response), with a CC to four internal employees:

Dear Heather

Many thanks for raising this issue with us. We are actually working on two fronts that will, we hope address your concerns in the near future (although not absolutely immediately). One is a general review of our policies, instructions and guidelines for the PLOS journals around data issues, and the other is improvements to the instructions and forms we use with reviewers. Both of these give us good opportunities to improve what we ask for and stipulate around data, which as you suggest is not yet optimal. It would be great if you let us know of other opportunities you think we’re missing, or any other suggestions you have in this area.



and I replied:

Thanks for the response, Theo!

This all sounds good, though I hope you don’t hold off on easy small improvements (adding a sentence or two to reviewer instructions to ask whether existing author instructions on data have been followed) until large changes are thoroughly designed and implemented.  [..]

Summary:  journals want to hear from us.  It is definitely worth the time to raise these issues. Please write to your journals too!