ANITA K. BAHN, SC.D.
This content is PDF only. Please click on the PDF icon to access.
To the editor: In the article, "Digital Capillary Blood Flow in Clubbing" (Ann Intern Med 75:933-935, 1971), Table 1 has the following footnote: "P = probability that difference between the means is caused by chance."
This is a common misconception that I believe should be corrected for your readers. Since the misstatement is very similar to one of the exercises in my textbook, "Basic Medical Statistics," I thought it might be instructive to quote this problem and the answer to it.
P = .01 means that the probability is one in a hundred that the observed difference is due
BAHN AK. P and the Null Hypothesis. Ann Intern Med. ;76:674. doi: 10.7326/0003-4819-76-4-674_1
Download citation file:
Published: Ann Intern Med. 1972;76(4):674.
Copyright © 2019 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use