Simply Statistics A statistics blog by Rafa Irizarry, Roger Peng, and Jeff Leek

Junior scientists - you don't have to publish in open access journals to be an open scientist.

By Jeff Leek

Editor’s note - This is a chapter from my book How to be a modern scientist where I talk about some of the tools and techniques that scientists have available to them now that they didn’t before.

Publishing - what should I do and why?

A modern scientific writing process goes as follows.

  1. You write a paper
  2. You post a preprint a. Everyone can read and comment
  3. You submit it to a journal
  4. It is peer reviewed privately
  5. The paper is accepted or rejected a. If rejected go back to step 2 and start over b. If accepted it will be published

You can take advantage of modern writing and publishing tools to handle several steps in the process.

Post preprints of your work

Once you have finished writing you paper, you want to share it with others. Historically, this involved submitting the paper to a journal, waiting for reviews, revising the paper, resubmitting, and eventually publishing it. There is now very little reason to wait that long for your paper to appear in print. Generally you can post a paper to a preprint server and have it appear in 1-2 days. This is a dramatic improvement on the weeks or months it takes for papers to appear in peer reviewed journals even under optimal conditions. There are several advantages to posting preprints.

  • Preprints establish precedence for your work so it reduces your risk of being scooped.
  • Preprints allow you to collect feedback on your work and improve it quickly.
  • Preprints can help you to get your work published in formal academic journals.
  • Preprints can get you attention and press for your work.
  • Preprints give junior scientists and other researchers gratification that helps them handle the stress and pressure of their first publications.

The last point is underappreciated and was first pointed out to me by Yoav Gilad It takes a really long time to write a scientific paper. For a student publishing their first paper, the first feedback they get is often (a) delayed by several months and (b) negative and in the form of a referee report. This can have a major impact on the motivation of those students to keep working on projects. Preprints allow students to have an immediate product they can point to as an accomplishment, allow them to get positive feedback along with constructive or negative feedback, and can ease the pain of difficult referee reports or rejections.

Choose the journal that maximizes your visibility

You should try to publish your work in the best journals for your field. There are a couple of reasons for this. First, being a scientist is both a calling and a career. To advance your career, you need visibilty among your scientific peers and among the scientists who will be judging you for grants and promotions. The best place to do this is by publishing in the top journals in your field. The important thing is to do your best to do good work and submit it to these journals, even if the results aren’t the most “sexy”. Don’t adapt your workflow to the journal, but don’t ignore the career implications either. Do this even if the journals are closed source. There are ways to make your work accessible and you will both raise your profile and disseminate your results to the broadest audience.

Share your work on social media

Academic journals are good for disseminating your work to the appropriate scientific community. As a modern scientist you have other avenues and other communities - like the general public - that you would like to reach with your work. Once your paper has been published in a preprint or in a journal, be sure to share your work through appropriate social media channels. This will also help you develop facility in coming up with one line or one figure that best describes what you think you have published so you can share it on social media sites like Twitter.

Preprints and criticism

See the section on scientific blogging for how to respond to criticism of your preprints online.

Publishing - what tools should I use?

Preprint servers

Here are a few preprint servers you can use.

  1. arXiv (free) - primarily takes math/physics/computer science papers. You can submit papers and they are reviewed and posted within a couple of days. It is important to note that once you submit a paper here, you can not take it down. But you can submit revisions to the paper which are tracked over time. This outlet is followed by a large number of journalists and scientists.
  2. biorXiv (free) - primarily takes biology focused papers. They are pretty strict about which categories you can submit to. You can submit papers and they are reviewed and posted within a couple of days. biorxiv also allows different versions of manuscripts, but some folks have had trouble with their versioning system, which can be a bit tricky for keeping your paper coordinated with your publication. bioXiv is pretty carefully followed by the biological and computational biology communities.
  3. Peerj (free) - takes a wide range of different types of papers. They will again review your preprint quickly and post it online. You can also post different versions of your manuscript with this system. This system is newer and so has fewer followers, you will need to do your own publicity if you publish your paper here.

Journal preprint policies

This list provides information on which journals accept papers that were first posted as preprints. However, you shouldn’t

Publishing - further tips and issues

Open vs. closed access

Once your paper has been posted to a preprint server you need to submit it for publication. There are a number of considerations you should keep in mind when submitting papers. One of these considerations is closed versus open access. Closed access journals do not require you to pay to submit or publish your paper. But then people who want to read your paper either need to pay or have a subscription to the journal in question.

There has been a strong push for open access journals over the last couple of decades. There are some very good reasons justifying this type of publishing including (a) moral arguments based on using public funding for research, (2) each of access to papers, and (3) benefits in terms of people being able to use your research. In general, most modern scientists want their work to be as widely accessible as possible. So modern scientists often opt for open access publishing.

Open access publishing does have a couple of disadvantages. First it is often expensive, with fees for publication ranging between $1,000 and $4,000 depending on the journal. Second, while science is often a calling, it is also a career. Sometimes the best journals in your field may be closed access. In general, one of the most important components of an academic career is being able to publish in journals that are read by a lot of people in your field so your work will be recognized and impactful.

However, modern systems make both closed and open access journals reasonable outlets.

Closed access + preprints

If the top journals in your field are closed access and you are a junior scientist then you should try to submit your papers there. But to make sure your papers are still widely accessible you can use preprints. First, you can submit a preprint before you submit the paper to the journal. Second, you can update the preprint to keep it current with the published version of your paper. This system allows you to make sure that your paper is read widely within your field, but also allows everyone to freely read the same paper. On your website, you can then link to both the published and preprint version of your paper.

Open access

If the top journal in your field is open access you can submit directly to that journal. Even if the journal is open access it makes sense to submit the paper as a preprint during the review process. You can then keep the preprint up-to-date, but this system has the advantage that the formally published version of your paper is also available for everyone to read without constraints.

Responding to referee comments

After your paper has been reviewed at an academic journal you will receive referee reports. If the paper has not been outright rejected, it is important to respond to the referee reports in a timely and direct manner. Referee reports are often maddening. There is little incentive for people to do a good job refereeing and the most qualified reviewers will likely be those with a conflict of interest.

The first thing to keep in mind is that the power in the refereeing process lies entirely with the editors and referees. The first thing to do when responding to referee reports is to eliminate the impulse to argue or respond with any kind of emotion. A step-by-step process for responding to referee reports is the following.

  1. Create a Google Doc. Put in all referee and editor comments in italics.
  2. Break the comments up into each discrete criticism or request.
  3. In bold respond to each comment. Begin each response with “On page xx we did yy to address this comment”
  4. Perform the analyses and experiments that you need to fill in the yy
  5. Edit the document to reflect all of the experiments that you have performed

By actively responding to each comment you will ensure you are responsive to the referees and give your paper the best chance of success. If a comment is incorrect or non-sensical, think about how you can edit the paper to remove this confusion.

Finishing

While I have advocated here for using preprints to disseminate your work, it is important to follow the process all the way through to completion. Responding to referee reports is drudgery and no one likes to do it. But in terms of career advancement preprints are almost entirely valueless until they are formally accepted for publication. It is critical to see all papers all the way through to the end of the publication cycle.

You aren’t done!

Publication of your paper is only the beginning of successfully disseminating your science. Once you have published the paper, it is important to use your social media, blog, and other resources to disseminate your results to the broadest audience possible. You will also give talks, discuss the paper with colleagues, and respond to requests for data and code. The most successful papers have a long half life and the responsibilities linger long after the paper is published. But the most successful scientists continue to stay on top of requests and respond to critiques long after their papers are published.

Note: Part of this chapter appeared in the Simply Statistics blog post: “Preprints are great, but post publication peer review isn’t ready for prime time”