If you look at my CV, you’ll see that I have 3 published peer-reviewed papers.
Let’s not dwell on how small that number seems in terms of the academic job market. Instead, let’s do dwell on how small that number seems in terms of substantial contribution to scientific knowledge, an outcome I care about at a much more lofty, less self-interested, level. It’s also a much grander, and more abstract, outcome – which is convenient.
It’s convenient because a) as an early-career researcher, I don’t expect myself to have made any super-major contributions to the canon (yet), and because b) it’s much easier to cognitively reframe my overall contributions – not just my papers – to actually be in line with this goal.
This is where open data comes in, and becomes awesome.
A little while ago I had the experience of reading a “call for papers” – some people were doing a meta-analysis on intention in moral judgment – and going ‘oh!’ I have published and unpublished data on that!
The published data comes from a paper in which my advisor and I were testing a particular prediction made by the “linguistic analogy” of morality (spoiler: it doesn’t work, and may be a bad theory). The unpublished data comes 1) from a study Simon and I have tried (and failed) to have published several times (see this page – it’s the cognitive load one), and 2) from the first two studies of my PhD thesis, which I’ve written up as a paper yet to be published (but only rejected twice so far, so I have hope!)
This data turned out to be just the kind of thing the meta-analysts were interested in. So, I went back to the original data files, summarized the results, and sent them off. It felt awesome. Finally, these data would be contributing to something beyond my file drawer! I can’t wait to see the results of the analysis, problems of meta-analyses notwithstanding.
Wait-wait-wait, I hear you say. What does this have to do with open data? This awesomeness could be achieved even within the current APA guidelines, where data has to be shared with “competent researchers” if they ask for it.
Yes, that is true in theory. But, in practice, it’s a lot easier to share your data if you have expected to be sharing it all along. The expectation of openness makes you label your variables more sensibly. It makes you keep track of your composite variables more thoroughly. It makes you save your data to some repository more permanently accessible than whatever laptop you happened to have been given by your department. All of these things mean that when that request to share your data arrives – whether directly or through a call for data – you will avoid the sinking feeling of “where… what name… ugh”, and instead experience the warm glow of “yes! I can contribute!” – which, believe me, is much much better.
Try it, and you’ll see.
Another benefit of open data is that you may get mentioned in a blog post by Daniel Lakens, about why blog posts are awesome because of open data (and other reasons). There is something a bit circular going on here, but I’m too jetlagged to untangle it…