What are valid reasons for opting out of open science, i.e. not to share research progress and output openly?

Question

Open Science in the sense of sharing research as it unfolds has been around for not much more than a decade, so the initial reasons for most researchers not to adopt it were somewhat trivial, e.g.

• not having heard about it
• not having thought about it
• not knowing what it means
• not knowing anyone who actually practices it
• not knowing how it would affect their research/ career/ community.

In recent years, the term - albeit not the practice - has found its way into mainstream research communication, so more researchers have actually heard and thought about it or know someone who has tried it out. With these developments, the reasons for not sharing are becoming more complex and worth exploring in more detail.

In doing so, I'd like us to

• collect reasons that are being put forward in response to questions about why some research is not shared
• group and structure these reasons
• analyze their validity
• explore underlying reasons that may contribute to not sharing but are not pointed out explicitly
• look into ways to addressing both the valid reasons and the invalid ones in suitable ways.

I am aware that this ticket may not be the best place to do all that, but I think it can serve as a way to coordinate activities in this space.

Comments

At the Open Science Barcamp on March 12, 2018, we ran a session dedicated to this topic. The notes from the session are at https://etherpad.wikimedia.org/p/oscibar2018_session13 (don't think a good archival copy exists, and the Internet Archive could not fetch it). They contain an incomplete list of reasons that have been put forward, but not yet an assessment of their validity or any of the other further steps outlined in the question. I'll try to turn these notes into further tickets here.

As a first follow-up to the session, a blog post by Ben Kaden explores these notes and combines them with insights from a research project about data sharing during PhD research. That blog post sits at https://libreas.wordpress.com/2018/03/13/forschungsdatenpublikationen/ and is archived at https://web.archive.org/web/20180315001519/https://libreas.wordpress.com/2018/03/13/forschungsdatenpublikationen/ .

---

I suggest to reframe this question (because most arguments are not valid, or easily argued against): "What are typical argument for not being open, and how to respond to them?". That could be a good collection of typical (counter)arguments that come up in discussions.
To keep topics manageable, maybe even separate threads for "arguments against open data", "arguments against open access", ...

---

I agree to split this up as you suggest. Quick update though that https://doi.org/10.1101/266627 has a Table 5 with "A. Reasons for sharing data" and "B. Reasons for not sharing data", as per results of a survey of neuroimaging researchers.

Answer 1

According to https://twitter.com/CaAl/status/966279936028958720 (archive link), there are no "valid reasons not to participate in open science practices". Dozens of comments, hundreds of retweets and thousands of likes indicate that he hit a nerve.

Comments

One good reason that prevents Open Science in some research projects is if personal data is involved, e. g. in psychological experiments. Privacy protection laws that protect human rights about their own data rank higher than Open Science.