Ian Hussey - The ERЯOR project: A three-prong effort to improve post publication critique and error detection - Perspectives on Scientific Error 2024
osf.io/ayfek/
By itself and as currently implemented, academic peer review is not up to the task of comprehensively detecting errors in scientific publications, at least in the field of psychology. Error detection typically requires more resources than available for peer review: there are simply too many manuscripts, and errors can come in too many forms. Post publication scrutiny and critique, for example of influential or controversial claims, represents an important parallel system of scientific verification. Unfortunately, published work is rarely checked for errors, likely because this behavior is poorly rewarded, there is a shortage of relevant tools, and little training in error detection is available. In this talk, I will discuss our efforts to improve each of these issues through the ERЯOR project (Estimating the Reliability and Reproducibility of Research). First, borrowing the concept from cybersecurity research, the ERЯOR project is the first large scale Bug Bounty program for psychological science. Published research findings are scrutinized for errors, with monetary payouts to the authors vs. the error checkers contingent on whether errors are found. Second, the ERROR project will produce resources and training materials. Existing tools are being collated and documented and an R package is currently under construction. Lastly, I will discuss the masters’ degree course in post publication error detection I teach, and the need for comparable courses at other institutions - echoing Dorothy Bishop’s recent call in her blog post “Defence against the dark arts: a proposal for a new MSc course”.
Негізгі бет Ғылым және технология Ian Hussey - The ERЯOR project: A three-prong effort to improve error detection - PoSE 2024
Пікірлер