A recent article in The Chronicle of Higher Education blames the increased number of published scientific articles on “an avalanche of low-quality research,” and claims that this avalanche is damaging academia. I agree that there are certainly problems associated with the large volume of published research, but is it truly the crisis that the article claims?
It becomes more difficult finding articles for a particular area of interest, as the arXiv feed for the limited area of high energy phenomenology dumps 20+ articles into my RSS reader every evening. This is not entirely unmanageable, as I mostly scan abstracts for anything related to my current work, and ignore the rest. Plus, it is 2010, and a search function can turn up a paper on any topic I desire. However, the number of articles makes it more difficult to keep abreast of more subfields, and tangentially-related, though helpful, articles in other fields go unnoticed. Cross-pollination of ideas has been essential throughout the history of science, but it is more difficult in this era of increased specialization. The increased number of papers in all areas can’t be helping matters.
The “arms race” aspect of publishing puts pressure on researchers, particularly young ones, to publish X papers in Y years to get a good postdoc, a tenure-track position, and finally to receive tenure. Apart from the occupational stress involved, the production of papers as “least publishable units” may make the arc of one’s research more difficult to follow. In addition, publication should ideally be more than just a resume-builder; it’s supposed to be an important part of the quest for knowledge. I’m sure “publish or perish” has made more than one scientific idealist into a cynical resume-padder.
Other problems with the sheer number of papers include the reviewing load for professors (wouldn’t know) and the environmental impact of all those paper copies of journals (which shouldn’t be necessary in the internet age). But the authors go further, saying that the reason for the profusion of articles is that there are too many bad papers getting published. If there were some way to reduce the number of bad papers, they claim, the above problems will be alleviated.
But what does it mean for a paper to be worthless? Are there papers that are really worth so little that they shouldn’t even be published? I will explore this question in a post coming later this week.