In areas such as medicine, “experimental results that are positive (i.e. showing a significant finding)” are reported differently from “results that are negative (i.e. supporting the null hypothesis) or inconclusive”. This is the most prominent example of a so-called publication bias.
Is there a publication bias in ethics, too?
In the last months I worked on the topic of responsibility for emissions. Is the producer of a good whose production involves emissions responsible for those emissions? Or is the consumer of the good responsible? Unfortunately, I arrived at messy results. Firstly, I concluded that the question has no easy answer. Secondly, I concluded that insofar as a tentative answer is possible, both the consumer and the producer are responsible. Thirdly, I concluded that it would be extremely difficult to determine shares of responsibility for the producer and the consumer.
How much neater would it have been, if I had arrived at a determinate and simple result! It would make presentation and publication of the results more attractive.
It’s not that I find my results boring. Given that in practice, we usually account emissions to the producer and given that some find it intuitively fairer to account emissions to the consumer, I did find it interesting to claim that the truth is, first of all, much less obvious than it seems and, secondly, probably somewhere in the middle.
Still: Even if my results are valuable and informative – in the same way that non-significant results in experimental studies are valuable and informative – they sell less well than clear-cut and simple results.
I wonder if this effect skews the publication of articles areas such as ethics, too, especially in applied ethics. In applied ethics, many issues must ultimately be evaluated as mixed bags and as inconclusive. Do papers that arrive at such disappointing conclusions get published as easily as papers who claim to have found certain and radical conclusions?
P.S.: Here’s a nice example of publication bias 🙂