Open data for the busy researcher

A PRO story: the value of creating a stub.

Today I got an email from an author who received a review from a PRO reviewer. The editor had not passed on the PRO reviewers’ request for details, and so the PRO reviewer submitted the suggested minimal review. The editor passed this review on to the author, who then asked me for advice on handling the situation. I thought my advice might be useful to others, so I’m outlining it here.

The author wanted to share their data, but they said that due to several factors they were time constrained. This is understandable; currently, psychological science is in a transition period where open data and materials are increasingly expected and valued, but journals still don’t have policies encouraging it. It is not something that is expected on submission, and so authors may not think about it until later. Hopefully this will change in the future, but this blog post is about the present. What should the author do?

It is important to emphasise that PRO is about transparency, not about forcing data sharing. Ideally, we’d like at least some data to be made easily (perhaps publicly) accessible to other researchers; but we also want to know when that is not possible, and why. If an author simply writes in the paper “The data will not be made public because…” and give any reason, that is enough for PRO. So, “The data will not be made public due to time constraints” and perhaps “the data will be made available on publication” are, if spelled out in the paper, totally fine responses for a PRO reviewer. (note that a reviewer might also want to see the data for verification during review; that is not covered by PRO, so I won’t deal with that here). A single sentence is all that is needed.

But in this case the author wanted to share their data. What to do? You want to take the proper amount of time to get the data in shape, etc; you also want the publication process to keep rolling. This is what I suggested.

• Create an OSF page for the project. If you’ve never used OSF before, you may have to create an account, but this is easy. If you need help, there are guides.
• Add a README file describing the project (perhaps title and abstract, and, importantly, your contact information). Say that the data/materials for the project will be placed there, and in the meantime, people who want access to the data/materials should contact you. This is a “stub” OSF page, awaiting your later additions.
• Add the link to this OSF page in your manuscript (perhaps in the author note). Write something like “Information about obtaining data and materials underlying this paper can be found at X.” where X is your OSF page’s URL.
• In your response to the reviews, tell the editor what you’ve done in response to the comment (maybe tell the editor you would have appreciated hearing from the PRO reviewer earlier :) ).
• When you have time — hopefully before the paper is formally published — populate the OSF page.

To do everything I’ve described above takes only a few minutes, at most, except for the last step. Consider the worst case scenario: you forget to upload the materials/data. Then a curious researcher simply emails you as would be typical, and asks you for it. The link to the OSF page points to your contact information, so nothing is lost.

But if you forget to upload the data/materials, the very first time someone asks you for them, simply prepare them as you typically would and then post them to the OSF page. Point the requester there. From then on, anyone who clicks on that link in your paper sees all the data and material. This is superior to the typical “data available on request” scenario, where you need to deal with every incoming email (assuming the data can be shared publicly, that is). In this case, you only have to deal with the first requester; the rest take care of themselves.

The best case scenario is that you get the data shared on your own timeline, but before the paper is published. In this case, the PRO reviewer is happy, you’re happy, and all the people that benefit from your materials and data are happy.