Statistical Graphics Using ODS


Principal Component Analysis Plots with PROC PRINCOMP

This example is taken from Example 91.3 in ChapterĀ 91: The PRINCOMP Procedure. The following statements create a SAS data set that contains job performance ratings of police officers:

options validvarname=any;

data Jobratings;
   input ('Communication Skills'n
          'Problem Solving'n
          'Learning Ability'n
          'Judgment Under Pressure'n
          'Observational Skills'n
          'Willingness to Confront Problems'n
          'Interest in People'n
          'Interpersonal Sensitivity'n
          'Desire for Self-Improvement'n
          'Appearance'n
          'Dependability'n
          'Physical Ability'n
          'Integrity'n
          'Overall Rating'n) (1.);
   datalines;
26838853879867
74758876857667

   ... more lines ...   

;

The following statements run PROC PRINCOMP:

proc princomp data=Jobratings(drop='Overall Rating'n) n=2
              plots=(Matrix PatternProfile);
run;

The plots are requested by the PLOTS=(MATRIX PATTERNPROFILE) option. The results, shown in FigureĀ 21.11, contain the default scree and variance-explained plots, along with a scatter plot matrix of component scores and a pattern profile plot.

Figure 21.11: Principal Component Analysis

Principal Component Analysis
External File:images/pcex1.png
External File:images/pcex2.png