🔢Contingency Table

NEW

Also known as: Cross Tabulation

A cross-tabulation summarizing the frequency distribution of variables for categorical data.

DESCRIPTION

A contingency table organizes counts of observations across categories of two or more variables, revealing associations or independence. Analysts compute expected counts and apply chi-square or Fisher’s exact tests to assess relationships. Marginal totals and proportions facilitate interpretation of joint and conditional distributions. In BI tools, crosstabs and pivot tables are practical implementations. Contingency tables underpin epidemiology (risk ratios, odds ratios) and marketing analytics (lift, response rates). Visualizations like mosaic plots complement tabular analysis.

ALTERNATIVE DEFINITIONS

Wikipedia

In statistics, a contingency table is a type of table in a matrix format that displays the frequency distribution of the variables.

Highlights matrix layout and frequency distribution for variables.

Source

NIST/SEMATECH e-Handbook

Contingency tables are used to examine relationships between categorical variables by displaying frequencies in a table format.

Emphasizes relational analysis of categorical variables.

Source

Penn State Eberly College of Science

A contingency table is a tabular summary of categorical data showing frequencies of observations for combinations of levels of two or more variables.

Focuses on multi-variable categorical combinations and frequency counts.

Source

RELATED TERMS

Chi-square Test Pivot Table Odds Ratio Fisher’s Exact Test Marginal Distribution Joint Distribution Risk Ratio Independence

COMPETING TERMS

Confusion Matrix

RELATIONS

• Broader: Table

KNOWLEDGE AREAS

Data Science Mathematics Data Visualization Data Warehousing & BI

TARGET AUDIENCE

Data Scientist Data Analyst BI Specialist Information Analyst

EXTERNAL SOURCES

• Wikipedia - Contingency table
• NIST - Statistical handbook entry
• Penn State - Course notes on crosstabs
• UCLA IDRE - Crosstab creation guide

TOP QUESTIONS

How is a Contingency Table used in hypothesis testing?

It provides observed counts to compute expected counts and the chi-square statistic, testing independence between categorical variables.

Category: Testing

What are common mistakes with Contingency Tables?

Using percentages without raw counts, ignoring sparse cells, and misinterpreting marginal distributions as causal relationships.

Category: Mistakes

How do you visualize a Contingency Table?

Use mosaic plots, stacked bar charts, and heatmaps to show joint distributions and highlight associations.

Category: Example

When should Fisher’s Exact Test be preferred?

When sample sizes are small or expected cell counts are low, Fisher’s test provides exact inference.

Category: Requirements

What is the role of marginal totals in a Contingency Table?

They represent totals per category and help compute conditional probabilities and expected counts.

Category: Overview

Can a Contingency Table handle more than two variables?

Yes, multi-way tables capture interactions among more variables but are harder to visualize and interpret.

Category: Overview

What is the risk of misinterpreting Contingency Tables?

Confounding and omitted variables can create spurious associations; correlation does not imply causation.

Category: Risks

How does a Confusion Matrix relate to a Contingency Table?

A confusion matrix is a special contingency table for predicted vs. actual classes in classification tasks.

Category: Example

Where to start learning Contingency Tables?

Study chi-square analysis from NIST or Penn State resources and practice with pivot tables and mosaic plots.

Category: Overview