Dhaliwal, S. and Campbell, M.J. (2010) Misinterpreting p-values in research. Australasian Medical Journal, 1 (1). pp. 1-2. ISSN 1836-1935
Abstract
The overuse of p-values to dichotomize the results of research studies as being either significant or non-significant has taken some investigators away from the main task of determining the size of the difference between groups and the precision with which it is measured. Presenting the results of research as statements such as “p < 0.05”, “p > 0.05”, “NS” or as precise p-values has the effect of oversimplifying study findings. Further information regarding the size of the difference between groups is required. Presenting confidence intervals for the difference in effect, of say two treatments, in addition to p-values, has the distinct advantage of presenting imprecision on the scale of the original measurement. A statistically significant test also does not imply that the observed difference is clinically important or meaningful, and their meanings are often confused.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | © 2010 Australasian Medical Journal. Reproduced in accordance with the publisher's self-archiving policy. |
Keywords: | p-value, confidence interval, clinical significance, equivalence test |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Medicine, Dentistry and Health (Sheffield) > School of Health and Related Research (Sheffield) |
Depositing User: | Miss Anthea Tucker |
Date Deposited: | 07 Sep 2010 10:48 |
Last Modified: | 05 Jun 2014 01:57 |
Published Version: | http://dx.doi.org/10.4066/AMJ.2009.191 |
Status: | Published |
Publisher: | The Australasian Medical Journal pty ltd. |
Refereed: | Yes |
Identification Number: | 10.4066/AMJ.2009.191 |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:11208 |