Behind the Façade of Self-Correcting Science

Abstract

Prior to my involvement in the hexagon case, I did not consider falsification to be a significant problem. The few times that the subject came up in conversation with colleagues, I can recall expressing my belief that no one would falsify data because they would be caught immediately. That is to say, I believed the myth of self-correcting science. That changed when I confronted people who were intentionally refusing to confront their falsifications and, indeed, threatening others in an attempt to squelch the truth. Here, I am not only referring to the respondents, but also university administrators. Many people rely on the idea of self-correction, saying ‘‘it will get sorted out in the journals.’’ If that were true, it would absolve the rest of us of any responsibility. When one actually tries to refute falsified data in a journal article, one realizes that the journals are not designed for that purpose. They are intended for new, fresh research. An author may not point out that other’s data are, in fact, misrepresented and not supported by the primary sources. Self-correction is a convenient myth that allows people to put an ugly reality out of their minds. However, the myth is brought up so frequently that it merits a spotlight to expose its absurdity.

Considering its central importance, one might expect replication to be a prominent part of scientific practice. It is not. An important reason for this is that scientists have strong incentives to introduce new ideas but weak incentives to confirm the validity of old ideas. Innovative findings produce rewards of publication, employment, and tenure; replicated findings produce a shrug.

– The Reproducibility Project: Psychology, Brian A. Nosek and dozens of authors, Perspectives on Psychological Science, 2012, 7, 657