Binomial 2×2 test for independent samples with independent proportions

1 January 1981

journal article
research article
Published by Taylor & Francis in Communications in Statistics - Theory and Methods

Vol. 10 (1) , 11-29
https://doi.org/10.1080/03610928108828016

Abstract

For two independent samples with independent proportions, the significance level of the difference between them can be obtainedfrom joint binomial probabilities, assuming the population propor-tion is well estimated by the samples' pooled proportion. An efficient computer program can be written for this binomial test. The x² test, using a continuity correction of .25 for equal sample sizes and .1 for unequal sample sizes, is in close agreement with the binomial test.

Keywords

SAMPLE PROPORTIONS
POPULATION PROPORTION
FISHER'S TEST
NORMAL TEST
CHI-SQUARE TEST
BARNARD'S TEST
CONTINUITY CORRECTION
COMPUTER PROGRAM

This publication has 2 references indexed in Scilit:

EXTENSION OF THE NEYMAN-PEARSON THEORY OF TESTS TO DISCONTINUOUS VARIATES
Biometrika, 1950
THE MEANING OF A SIGNIFICANCE LEVEL
Biometrika, 1947