[Skip to Navigation]
May 9, 2019

Use of Confidence Intervals to Improve the Quality of Research Results Reporting and Interpretation

Author Affiliations
  • 1Department of Otolaryngology–Head & Neck Surgery, Washington University in St Louis School of Medicine, St Louis, Missouri
  • 2Statistics Editor, JAMA Otolaryngology–Head & Neck Surgery
  • 3Department of Otolaryngology–Head and Neck Surgery, University of Wisconsin School of Medicine and Public Health, Madison
  • 4Endoscopy and Laryngology Editor, JAMA Otolaryngology–Head & Neck Surgery
  • 5Editor, JAMA Otolaryngology–Head & Neck Surgery
JAMA Otolaryngol Head Neck Surg. 2019;145(6):491-493. doi:10.1001/jamaoto.2019.0793

Theoretical principles underpinning the P value and hypothesis testing were conceived by Fisher (P value)1 and Neyman and Pearson (null hypothesis testing)2 in the 1920s and 1930s. The originators of these concepts had no idea that their probabilistic theories would become the reference standard for scientific discovery. In fact, these same pioneers argued that “no test based upon the theory of probability can by itself provide any valuable evidence of the truth or falsehood of that hypothesis.”2(291) Nonetheless, P values and the null hypothesis testing paradigm and have become ubiquitously associated with scientific discovery without careful consideration of what they mean.

Add or change institution