[Skip to Content]
[Skip to Content Landing]
Other Articles
January 21, 1961

A Note on Statistical Usage

JAMA. 1961;175(3):257. doi:10.1001/jama.1961.03040030081025

This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.


To the Editor:—  In reply to your letter of July 21, I have now had an opportunity to discuss Dr. Hawkins' paper with my associate, Dr. Norbert Freedman.Dr. Hawkins is correct in pointing out an ambiguity in the interpretation of our data (Englehardt and others: Prevention of Psychiatric Hospitalization with Use of Psychopharmacological Agents, JAMA173:147-149 [May 14] 1960). Chlorpromazine and promazine are not homogeneous in their treatment effects, insofar as we assessed their treatment effects. The data do not warrant a conclusion regarding a combined chemotherapeutic effect, and no such interpretation was intended by us even though the paper might have created this impression.These considerations do not invalidate our major conclusion. The Table in our paper giving frequencies for 3 treatment conditions and 2 types of outcome showed a non-chance association. Type of treatment was a significant source of variation affecting treatment outcome. Inspection of the

First Page Preview View Large
First page PDF preview
First page PDF preview