What’s the cure for an academic epidemic?

Ella Rhodes reports.

A position paper that outlined five ‘diseases’ prevalent in scientific publishing has been catching the attention of scores of academics. John Antonakis, a psychologist by training, writing as new editor of The Leadership Quarterly, succinctly outlined the problems with academic publishing, often discussed within psychology, and what can be done by publishers to combat these.

The problems in academic publishing can’t be wished away; and while it’s often said that science is self-correcting, Antonakis argued this won’t be the case if publishers aren’t properly managing the process of publication. He added: ‘For whatever reasons, including failure to robustly replicate research, there are simply too many findings that become “received doctrines”…
or “unchallenged fallacies”.’

The very system of academic publishing is causing issues; where academics aim for the highest-impact journals, which in turn tend towards publishing novel, sparkling and significant findings (particularly in psychology), the most robust or policy-relevant findings aren’t necessarily seeing the light of day. Antonakis wrote that although he can’t give a definitive diagnosis of the broader issues in academic publishing, a likely explanation is to do with the conditions and incentives in place to reward success in the production of research.
Antonakis outlined the five ‘diseases’ that are stifling the publication of useful research: significosis, neophilia, theorrhea, arigorium and disjunctivitis. The first refers to the tendency for statistically significant research to be published over that with non-significant results; the second is a propensity for novel or innovative research to be published often ignoring null results or replications. Theorrhea, Antonakis described as a ‘mania for new theory’; arigorium refers to a lack of rigour in theoretical and empirical work; and disjunctivitis is the propensity to produce high volumes of ‘redundant, trivial and incoherent works’.

These problems are probably familiar to many in the academic world, particularly within the social sciences, and questions about the best approach to tackle them have been tossed back and forth for years. Antonakis makes a number of recommendations to change this status quo, changing publication incentive structures and our ways of thinking about what makes a useful contribution to a given field. Indeed, as editor of The Leadership Quarterly Antonakis will now accept a broader range of articles, including replication and null-results studies, and registered reports, where researchers log their experimental methods and hypotheses with journals prior to carrying out the research. er

To read Antonakis’ full paper, ‘On doing better science: From thrill of discovery to policy implications’, see
tinyurl.com/jk6scek

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber