How to use χ2 test correctly——the overview of the statistical analysis methods for the data of a multiway table

Abstract

The purpose of this article was to introduce the types of a multiway table data and their corresponding statistical analysis methods. Based on the data types of the result variables， common high-dimensional table data could be divided into the following three categories， namely， the high-dimensional table data with binary variables， multi-valued nominal variables and multi-valued ordinal variables as the result variables. There were two main categories of the statistical analysis methods for the multiway table data. The first category was the generalized difference analysis， in which the contents included "weighted χ2 test""CMH χ2 test" and "Meta analysis"， the second category was the regression analysis， in which the contents included "log linear regression model analysis""Logistic regression model analysis" "probit regression model analysis" and "discrete choice model analysis".

Keywords

• weighted χ2 test
• cmh χ2 test
• meta analysis
• discrete choice model