Basic Analysis > Contingency Analysis
Publication date: 07/08/2024

Contingency Analysis

Examine Relationships between Two Categorical Variables

Use the Contingency platform to investigate the relationship between two categorical variables. The categorical variables can be ordinal or nominal. The analysis results include a mosaic plot, a contingency table of frequency counts and proportions, and chi-square tests of significance. You can interactively perform additional analyses and tests on your data, such as analysis of means for proportions, correspondence analysis, and measures of association.

The Contingency platform is the categorical by categorical personality of the Fit Y by X platform.

Figure 7.1 Example of Contingency Analysis 

Example of Contingency Analysis

Contents

Example of Contingency Analysis

Launch the Contingency Platform

Data Format

The Contingency Report

Mosaic Plot
Contingency Table
Tests Report

Contingency Platform Options

Contingency Analysis Reports

Analysis of Means for Proportions Report
Correspondence Analysis Report
Cochran-Mantel-Haenszel Test Report
Agreement Statistic Report
Relative Risk Report
Two Sample Test for Proportions Report
Measures of Association Report
Cochran Armitage Trend Test Report
Fisher’s Exact Test Report
Equivalence Test Reports

Additional Examples of the Contingency Platform

Example of Analysis of Means for Proportions
Example of Correspondence Analysis
Example of a Cochran-Mantel-Haenszel Test
Example of the Agreement Statistic Option
Example of the Relative Risk Option
Example of a Two Sample Test for Proportions
Example of the Measures of Association Option
Example of the Cochran Armitage Trend Test
Example of Equivalence Tests for Relative Risk

Statistical Details for the Contingency Platform

Statistical Details for the Agreement Statistic
Statistical Details for the Odds Ratio
Statistical Details for the Tests Report
Statistical Details for Correspondence Analysis
Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).