📊

Creating Clustered Box Plots in SPSS

Sep 26, 2025

Overview

This lecture explains how to create clustered box plots in SPSS using different methods, depending on whether data is separated by category columns or organized with a category variable.

Creating Clustered Box Plots with Category Variable

  • Open Graphs > Chart Builder in SPSS to create box plots.
  • Choose "Simple Box Plot" and place the measurement variable (e.g., age) on the Y-axis, category variable (e.g., gender) on the X-axis.
  • SPSS displays side-by-side box plots for each category (e.g., male and female).
  • The "Clustered Box Plot" option groups categories, but you may need to adjust the clustering variable settings to enable the OK button.

Using Box Plot Menu for Separate Variables

  • Go to Graphs > Boxplot, select "Simple", then "Summaries of separate variables".
  • Add each category (e.g., male, female) as separate variables to display their box plots side by side.
  • Note: Box plot scaling may differ depending on whether axis starts at zero or the minimum data value.

Using Analyze > Descriptive Statistics > Explore

  • Navigate to Analyze > Descriptive Statistics > Explore for an alternative approach.
  • Add the measurement variable to "Dependent List" and the category variable to "Factor List".
  • Under "Plots", select only "Boxplot" and deselect other plot types.
  • This method outputs both descriptive statistics and box plots by category.

Handling Data in Separate Columns

  • If data for each category is in a separate column, use Graphs > Boxplot > Simple > Summaries of separate variables.
  • Select both category columns (e.g., male and female) as dependent variables to display box plots.
  • For Analyze > Explore, add both category columns to the "Dependent List" for generating box plots.

Key Terms & Definitions

  • Clustered Box Plot — A graphical display showing the distribution of a variable across multiple categories side by side.
  • Category Variable — A variable indicating the group (e.g., gender) to which each data point belongs.
  • Outlier — A data value significantly different from others in the dataset, often marked on box plots.

Action Items / Next Steps

  • Practice making clustered box plots in SPSS using both category variable and separate columns.
  • Review the differences in axis scaling when interpreting box plots.
  • Try Explore and Chart Builder methods to compare outputs.