Chapter 6
Inferences Involving Binomial and Nominal
Count Data
One-Sample Tests (two measures)^
6.1 One-sample χ
(^2) test of independence
165
One-sample test for binomial or multi-category data in the form of
frequency counts. Used to test the significance of an association
between two categorical variables.6.1 Phi Coefficient 174
A descriptive measure of the strength of association between two
categorical variables in a 2×2 table. Often used in conjunction with a
χ^2 test.6.1 Cramer’s Phi (sometimes called Cramer’s V) 174
A descriptive measure of the strength of association between two
categorical variables when there are more than four cells in the
contingency table.6.2 Binomial test 175
A one-sample test that is useful for comparing two proportions or
percentages. The two percentages should add up to 100 per cent
(because this is a one-sample test).Two-Sample Tests^
6.1 r×2 Sample χ
(^2) test of homogeneity
165
A test of homogeneity of proportions in a contingency table. When
subjects are sampled from two separate populations, this test is used
to compare the distribution of proportions in the two independent
populations.