DAVID W. FRASER, M.D.
This content is PDF only. Please click on the PDF icon to access.
To the editor: I wish to point out the statistical errors in the paper by Drs. Comty, Cohen, and Shapiro on uremic pericarditis in the August issue (Ann Intern Med 75:173-183, 1971).
The authors point out that their chi-square manipulations "cannot be taken at face value." In fact, the chi-square test is inappropriate for data with such small observed and expected values; the Fisher's exact test is suitable.
None of the individual entries nor subtotals recorded in their Table 8 as "expected" is correct. In addition, the subtotals of "expected" add up to 22, not 25.
I am made uncomfortable
FRASER DW. Statistics on Uremic Pericarditis. Ann Intern Med. 1971;75:976. doi: https://doi.org/10.7326/0003-4819-75-6-976_1
Download citation file:
Published: Ann Intern Med. 1971;75(6):976.
Copyright © 2020 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use