Everything You Wanted to Know but Could Never Find from the Cochran-Mantel-Haenszel Test

The Cochran-Mantel-Haenszel (CMH) test, developed in the 1950s, is a classic health research, especially epidemiology and other fields which dichotomous polytomous variables are frequent. This nonparametric test makes it possible to measure check effect of an antecedent variable X on outcome Y, statistically controlling third Z that acts as confounding relationship between Y. Both Y measured qualitative scale polytomous-qualitative or ordinal scale. It assumed homogeneous k strata Z, usually tested by Breslow-Day with Tarone’s correction Woolf’s test. main statistical programs have CMH together verify assumption across strata, so easy apply. However, its fundamentals details calculations mystery most researchers, even difficult find understand. aim this article present these clear concise way, including assumptions alternatives non-compliance. technical knowledge applied simulated realistic example area and, finally, interpretive synthesis analyses given. In addition, some suggestions for report made.