Crosstab analysis of two categorical variables

exploratory statistics

slug: recipe-exploratory-statistics-crosstab-analysis-of-two-categorical-variables

@@## Recipe: Crosstab analysis of two categorical variables category: exploratory statistics

Problem

You need to examine relationships between two categorical variables:

• understand joint distributions (e.g., product category vs. customer segment)
• detect patterns or dependencies
• identify unexpected combinations or data quality issues

Solution

Follow these steps to perform a two-way frequency analysis:

• load the dataset
• select two categorical variables
• compute a cross-tabulation of counts or percentages
• optionally visualize as a heatmap or table

Step Sequence

load step -> [step-xfreq] -> filter step -> chart step

Input Datasets

• transactions_clean — cleaned transactional data
• Notes: select categorical fields such as product_category and customer_segment

Output Dataset

• crosstab_table — table showing counts or percentages for each combination of the two variables
• Notes: can be used to detect trends, dependencies, or anomalies

Step-By-Step Explanation

Step	Purpose	Notes
load step	Load dataset	Supports local file, database, or API sources
[step-xfreq]	Compute cross-tabulation	Example: count transactions per product_category × customer_segment
filter step	Optionally exclude rare combinations	Example: combinations with <5 occurrences
chart step	Visualize cross-tabulation	Optional heatmap, stacked bar chart, or table

Variations & Extensions

• Apply calculate step to compute percentages or normalized values
• Combine with compare step to compare crosstabs across datasets or periods
• Include dashboard step to create interactive views of joint distributions

Concepts Demonstrated

• Crosstab / two-way frequency analysis
• Identifying dependencies between categorical variables
• Highlighting rare or unexpected combinations
• Sequencing statistics and visualization steps

Related Recipes

• Frequency analysis of categorical data
• Univariate analysis of numeric variables

Notes & Best Practices

• Inspect low-frequency combinations for potential data issues
• Use percentages to standardize across different dataset sizes
• Document the variables used and any filtering applied

Metadata

title: "Crosstab analysis of two categorical variables"
category: "exploratory statistics"
difficulty: "Intermediate"
tags: [crosstab, xfreq, categorical, EDA]
inputs: [transactions_clean]
outputs: [crosstab_table]
steps: [step-load, step-xfreq, step-filter, step-chart]
author: "Tom Argiro"
last_updated: "2025-10-25"
doc_type: "recipe"