,
7 tweets,
2 min read
Read on Twitter
Our publication system is broken. When you are evaluated based on how many papers you publish, scientists respond with lots and lots of papers that represent incremental advances (at best). The result is a lot of useless papers clogging the peer-reveiwed literature. 1/
This has many bad knock-on effects. 1) it means that actual, legitimate advances in science can be buried without notice (esp. if the advance is by a grad student/postdoc who is unknown to most people). 2/
2) because there are so many papers published, it's become nearly impossible to know everything that's been done on a subject, so papers frequently duplicate work that's already been done. 3/
3) the sheer volume of papers is wearing out peer reviewers. it can be hard to find qualified reviewers, and those that do agree to review simply can't spend as much time as they should reviewing papers, leading to a feedback that further lowers standards for publication. 4/
Ultimately, the solution is to evaluate people on the quality of publications, not on the number. I'm not sure how to do that. As one colleague jokes, Department Head and Deans only know how to count to ten — twenty if they use their toes. 5/
But I do know this: we're poised on the brink of an explosion of science from countries like China, that are rapidly spooling up their international research profile. This problem is going to get worse before it gets better. 6/
I’d be curious to know what @steveghan1 thinks. Is there a way to make the literature more useful for the sci community?
