SUSAN ELLENBERG, PH.D.; JANET WITTES, PH.D.
This content is PDF only. Please click on the PDF icon to access.
To the editor: In an indictment of classical statistical methodology, Diamond and Forrester (1) have proposed an alternative Bayesian methodology for analyzing data from clinical trials. In a subsequent correction, they pointed out a mislabeled figure and wrote that the text associated with the figure ". . . is therefore, incorrect and should be disregarded" (2). We submit that their paper is so seriously flawed that it should be disregarded in its entirety.
The authors claim their figures and arguments show that highly significant results seen with small sample sizes, as well as conventionally significant results seen from large sample
ELLENBERG S, WITTES J. Analysis of Data from Clinical Trials. Ann Intern Med. ;99:874–875. doi: 10.7326/0003-4819-99-6-874
Download citation file:
© 2019
Published: Ann Intern Med. 1983;99(6):874-875.
DOI: 10.7326/0003-4819-99-6-874