Hostname: page-component-7b7567c8c-z4zkx Total loading time: 0 Render date: 2026-11-18T07:01:26.910Z Has data issue: false hasContentIssue false

Comparisons of Power for Some Exact Multinomial Significance Tests

Published online by Cambridge University Press:  01 January 2025

Joan M. Gurian
Affiliation:
National Heart Institute
Jerome Cornfield
Affiliation:
National Heart Institute
James E. Mosimann
Affiliation:
National Institute of Neurological Diseases and Blindness National Institutes of Health

Abstract

Two small-sample tests (proposed by Tate and Clelland and by Chapanis respectively) of hypotheses about the parameters of the multinomial distribution, where

\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$f(x|p) = n!\prod\limits_{i = 1}^k {\frac{{p_i^{x_i } }}{{x_i !}}} $$\end{document}
are described. A proof is given that a test of the form
\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$R = \frac{{\smallint f(x|p)g(p)dp}}{{f(x|p_0 )}}$$\end{document}
, where g(p) ≥ 0 and p0 is the vector of p's specified by the hypothesis, is admissible. It is shown that the Tate and Clelland test is obtained by setting g(p) equal to a constant for all p and is therefore admissible. The Chapanis test is shown to have power less than or equal to the power of the Tate and Clelland test for (k = n = 3) and (k = 3, n = 4).

Information

Type
Original Paper
Copyright
Copyright © 1964 Psychometric Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable