In the Contingency platform, the Cochran-Mantel-Haenszel test evaluates the relationship between two categorical variables after blocking across a third classification. Blocking across a third classification variable is also called stratification.
Note: See Example of a Cochran-Mantel-Haenszel Test.
The Cochran-Mantel-Haenszel Tests report contains a table with the results of various tests. Each test has a chi-square statistic (ChiSquare), associated degrees of freedom (DF), and significance probability (Prob>Chisq). The following tests are reported:
Correlation of Scores
(Applicable when both the Y and X variables are ordinal or interval.) Tests the alternative hypothesis that there is a linear association between the Y and X variables for at least one level of the blocking variable.
Row Score by Col Categories
(Applicable when the Y variable is ordinal or interval.) Tests the alternative hypothesis that, for at least one level of the blocking variable, the mean scores of the rows are unequal.
Col Score by Row Categories
(Applicable when X is ordinal or interval.) Tests the alternative hypothesis that, for at least one level of the blocking variable, the mean scores of the columns are unequal.
General Assoc. of Categories
Tests that for at least one level of the blocking variable, there is some type of association between the Y and X variables.
Tip: You can perform a row means score test by specifying a column that contains a constant value as the blocking variable in the Cochran-Mantel-Haenszel test.