##Sharing data
Goals for this lesson
- Understand the key motivators for why people do (and do not) share their data
- Understand the risks of data sharing
- Develop ideas for mitigating risk
Data or it didn't happen. Right?
So far, in this course, we've explored lots of ways to share data, learned the best practices around data sharing, and helped you become better data sharers. Frankly, I've painted a very idealistic view of data sharing. But should all data be shared? I can think of some pretty obvious cases where sharing data without any controls would be downright unethical- for example, personal health records of people who did not give permission for the data to be shared.
An example from ecology- imagine you were a scientist who collected data on a rare orchid, including location data. Sharing that data freely could make an already vulnerable species subject to greater risk of poaching. What sort of controls should we use in the sharing of those data?
Then, there's the issues surrounding data sharing at professional or personal cost to scientists. Most data sharing regulations require data to be shared upon publication of results, or after a short embargo period. This is based on the assumption that after publication, the work is done- but this is not necessarily the case. The data sharing quandry that I personally think about the most, in fact, is the one faced by scientists involved in long term studies, because we publish essentially interim reports, but the experiments keep going. This has the potential to expose scientists with long-term projects to greater risks as outlined in our discussion paper for today.
Resources:
- Archiving Primary Data: Solutions for Long-Term Studies by Mills et al- this paper represents a serious pushback against data sharing- I disagree with many of their arguments as an LTER scientist, but I admit I come at long term studies from a place with considerable privilege and probably an unusual reward structure because of my connection to this network. I feel this paper is an ideal nucleation point for discussion. Does data sharing come from a place of privilege? How can we make it not?