This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
The stress in this book is implied by the title; it is a book on experiment design and not a statistics book. The first 59 pages are devoted to the description of the problems which arise in the planning of medical experiments and in the type of observations to be made. Approximately 20 pages are devoted to the technic of calculation, the central point being the development of a "result-statistic," which is a constant to be referred to a table presenting probabilities for explaining observed results on the basis of chance. The remainder of the book is devoted to an analysis of the problems which have arisen in medical research, the majority having been taken from The Journal of the American Medical Association.
A minimum of mathematics is required of the reader. Only the simplest statistical tools are explained, the more complicated subjects, such as correlation, are left to other
Experiment Design and Judgment of Evidence. JAMA. 1948;138(14):1067. doi:10.1001/jama.1948.02900140059038