Data Sharing

What should libraries know about data sharing trends and what should we be doing differently to adapt along with this evolving landscape?

The march toward open science and data sharing continues – Oct 1, 2014 is the date after which the US Dept of Energy funded research must conform to new data management and data sharing requirements.

Even though I’m not directly connected with life sciences, I pored over a recent article in PLoS * while I reflected on that question.   The authors surveyed life science researchers to determine the kinds of influences that promote or constrain data sharing, and how life science researchers themselves perceive influences on their own data sharing practices.datasharing

I found this comparison interesting — 65% of survey respondents overall found NIH policies as positively influencing sharing, and 39% were positively influenced to share data as a result of formal instruction.  You would expect the NIH influence to be high — after all, that’s how researchers get funded and discussions about these national policies are widely discussed. For that reason I would expect the formal instruction percentage to be lower in comparison with the NIH figure.  In fact, that formal instruction has a positive effect on research sharing 39% of the time surprised me – it seems like a relatively strong positive response and makes me wonder what is going on in that formal instruction.  And, given that formal instruction is hard to scale, I wonder about the origin of instruction for researchers who are engaging in it. Some comes from libraries, I know, but where else are researchers going for the info they need? And how does library-based instruction compare with the other instruction researchers are getting (at conferences? in discipline-based programs?) and is there room for more coordination and collaboration?  What do researchers most need to know in making decisions about data sharing?  For the 57% for whom formal instruction did not influence data sharing, or the 4% who reported it had a negative influence on their data sharing, why was that?

Part of the answer to formal training having a negative effect is described in the study — there are institution-based technology and material transfer agreements that impede willingness to share on many levels.  Formal instruction may be informing researchers of requirements that seem onerous.

Life science researchers, like anybody else I suppose, base their communication practices — in this case data sharing — on social values that predicate sharing on mutual behavior. If a scientist doesn’t share their information with others or seems excessively self-interested, in return colleagues will refrain from sharing information with that person. In addition, there are the usual tradeoffs about protecting your own career status by withholding data until you can reap expected benefits from your research.

Other key factors in sharing data are well discussed in the article  – including the infrastructure available via open data repositories, the bureaucratic costs of complying with policies and guidelines, the low level of consequences for non-compliance in various ways, and so on.

The “getting scooped” problem is one I’m fairly sure libraries can play a role in, via researcher networks and institutional repositories that can help researchers publicize and report on their research as a work in progress, thereby kind of staking a claim and relying on social norms to keep others from free riding.  No one really wants to spend their dissertation time doing research someone else is already doing, so openness platforms designed to help prevent duplication of effort and prevent scooping/free riding would be a good service to researchers.  Another fruitful area would seem to be in the design of repositories so that metadata schemes can be extended as data sets are re-purposed.  And of course developing tools like the DMP Tool, that help provide researchers with information they need to do their work and reduce the bureaucratic cost of compliance is another kind of information service libraries can provide.

* Pham-Kanter G, Zinner DE, Campbell EG (2014) Codifying Collegiality: Recent Developments in Data Sharing Policy in the Life Sciences. PLoS ONE 9(9): e108451.
doi: 10.1371/journal.pone.0108451

Advertisements

About Karrie Peterson

Head, Liaison, Instruction & Reference Services
This entry was posted in Academic Technology Planning, Data Services, Organizational Effectiveness and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s