[Skip to Navigation]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
February 1967

Psychiatric Drug Research: Sample Size Requirements for One vs Two Raters

Author Affiliations

Galveston, Tex; Palo Alto, Calif; Galveston, Tex
From the Department of Neurology and Psychiatry, University of Texas Medical Branch, Galveston (Dr. Overall); the Veterans Administration Hospital, Palo Alto, Calif (Dr. Hollister); and the Research Computation Center, University of Texas Medical Branch, Galveston (Mr. Dalal).

Arch Gen Psychiatry. 1967;16(2):152-161. doi:10.1001/archpsyc.1967.01730200020004

IN THE evaluation of psychiatric drug treatments, two types of errors are sometimes present in conclusions based upon statistical analyses of data. It can be concluded that the treatments differ when, in fact, they do not; or it can be concluded that the treatments do not differ when, in fact, they do. In the past, most concern in the analysis of such data has been with the possibility of erroneously concluding that the treatments differ when they do not. Too little attention has been paid to the power of tests in detecting true treatment differences where they exist. When sample sizes are too small, statistical tests will fail to indicate significant differences between drug treatments, even though real and important differences exist.

The purpose of this article is to provide estimates of sample size required for doubleblind controlled studies using one, or two independent