The replication crisis

• Published in: Journal of advanced nursing Vol. 77; no. 2; pp. 501 - 503
• Main Author: Lipscomb, Martin
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.02.2021
• Subjects: Data replication; Medical research; Psychological research; Psychology; Research methodology; Social sciences; Statistical analysis
• ISSN: 0309-2402; 1365-2648; 1365-2648
• DOI: 10.1111/jan.14697

The problem is easily stated and well known. Research findings that cannot be reproduced or replicated lack trustworthiness and that's why, in part, reports detail the methods employed in finding derivation. Stating methods, including methods of statistical analysis, enables researchers to rerun and thereby test (confirm or not) the conclusions claimed in studies. However, it is now realized that many findings in, for example, psychology, social science, and some branches of medicine cannot be replicated (Stupple, Singerman, & Celi, 2019; Yong, 2018). The significance and ramifications of this realization are difficult to underplay. Findings that refuse replication undercut what we think we know and although replication might, as a topic, seem arcane, the subject is not simply of concern to those entangled in academic or research methods disputes. Popular science texts such as Tim Harford's How to Make the World Add Up (2020), and – in the UK – radio programmes such as Analysis’ The Replication Crisis (2018) have brought the issue to public attention. The subject even has its own Wikipedia page (https://en.wikipedia.org/wiki/Replication_crisis).