When you create categorized box plots, a series of standard 2D box plots, one for each category of cases identified by the X or X and Y category variables (or identified by the multiple subset criteria) is produced.

The categorized box and whisker plot typically summarizes the distribution of a variable broken down (categorized) by some other grouping or independent variable(s), by three components:

1. A central line to indicate central tendency or location;

2. A box to indicate variability around this central tendency;

3. Whiskers around the box to indicate the range of the variable (or another measure of variability).

One, two, or three grouping variables can be represented on one graph (i.e., a graph can represent a one-way, two-way, or three-way categorization of cases). If two or three grouping variables are selected, a "crosstabulation of graphs" is produced. These graphs allow the user to plot multiple dependent variables in one graph, support different graph layout formats, and offer a selection of categorization methods (including user-defined intervals, multiple subsets defined by logical expressions, and others).

In the categorized box and whisker plots called from the Results dialogs, the categorization (i.e., the breakdown into subsets of cases) based on the grouping variables uses the grouping codes (e.g., MALE, FEMALE) used in the current analysis (e.g., ANOVA or Breakdowns). If no grouping codes are used in the analysis, the program will ask for the codes before the graph is produced.

Usually, you can choose to plot for each categorized variable:

(1) Medians (central line), quartiles (box), and ranges (whiskers);

(2) Means, standard errors of the means, and standard deviations;

(3) Means, standard deviations, and 1.96 times the standard deviations (95% normal confidence interval for individual observations around the mean);

(4) Means, standard errors of the means, and 1.96 times the standard errors of the means (95% normal confidence interval for means).

One-way categorized box and whisker plots are available directly from all spreadsheets by using the Graphs of Input Data option (see Graphs of Input Data - Box-Whisker by...). In order to produce box plots representing the distribution of the values in cases (and not variables), use either the Graphs of Block Data - Rows option from the spreadsheet shortcut menu (instead of transposing your data file).

A wide variety of categorized box and whisker plots are also available from the Categorized Graphs options (in the Graphs menu, see Graphs menu graphs). These graphs allow the user to plot multiple dependent variables in one graph, support different graph layout formats, and offer a selection of categorization methods (including user-defined intervals, multiple subsets defined by logical expressions, and others). For a general overview of the graphics facilities in STATISTICA, see Graphs in STATISTICA.

See also, Scatterplots w/Box Plots, 2D Box Plots, and Graphs - Categorized Box Plots.