JBUON 2020; 25(2): 588-593 ISSN: 1107-0625, online ISSN: 2241-6293 • www.jbuon.com Email: editorial_ofce@jbuon.com OPINION ARTICLE Corresponding author: Constantinos E.Aliferis, MD. Icahn School of Medicine at Mount Sinai, 1468 Madison Ave, Annenberg Building 20 th foor, New York, NY, 10029, USA. Tel: +1 917 459-2620, Email: konstantinos.aliferis@mssm.edu; aliferis.k@gmail.com Received: 03/01/2020; Accepted:11/01/2020 The arbitrary magic of p<0.05: Beyond statistics Constantinos E. Aliferis 1 , Eleni Souferi-Chronopoulou 2,3 , Dimitrios T. Trafalis 4 , Antonios Arvelakis 1 1 Recanati/Miller Transplantation Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 2 Department of Pathology, Athens Medical School, National and Kapodistrian University of Athens, Athens, Greece. 3 Department of Statistics, University of Athens, Athens, Greece. 4 Department of Pharmacology–Clinical Pharmacology Unit, Athens Medical School, National and Kapodistrian University of Athens, Athens, Greece. Summary Purpose: Modern research and scientifc conclusions are widely regarded as valid when the study design and analysis are interpreted correctly. P-value is considered to be the most commonly used method to provide a dichotomy between true and false data in evidence-based medicine. However, many authors, reviewers and editors may be unfamiliar with the true defnition and correct interpretation of this number. This article intends to point out how misunderstanding or misuse of this value can have an impact in both the scientifc community as well as the society we live in. The foundation of the medical education system rewards the abundance of scientifc papers rather than the careful search of the truth. Appropriate research ethics should be practised in all stages of the publication process. Key words: statistics, medical reversal, biostatistics, ethics Introduction Most researchers feel that it is useless to sub- mit any paper for publication that lacks results of statistical signifcance and this concern is not ill-founded since most journals, chief editors and peer reviewers rely on the results of analyses that indicate a meaningful, impactful research article which can therefore be published. Scientists are pre-occupied in the focus of producing a p-value of less than 0.05. Signifcant or not? A real struggle. In statistics, one rule did we cherish: P point oh fve we publish, else perish! Said Val Johnson, “that’s out of date, our studies don’t replicate P point oh oh fve, then null is rubbish This limerick by the famous biostatistician Professor Roderick Little from the University of Michigan comes to underly the reality; research which produces p-values that achieve to surpass the arbitrary 0.05 is more likely to be published than research that does not. Studies that were nev- er published due to this limitation may have had equal or greater scientifc importance but remained unseen. On the other hand, this misuse of p-values can lead to false conclusions. Origins The search for tests of statistical signifcance began early in the history of statistics. In 1893 Pearson described the χ 2 test and while presented various results, the following comments came from his well-known paper [1]: “p= .1 (not very improb- able that the observed frequencies are, compatible This work by JBUON is licensed under a Creative Commons Attribution 4.0 International License.