CARDINALITY Procedure

(View the complete code for this example.)

This example illustrates how you can use PROC CARDINALITY to determine the limited cardinality of a variable. It uses the same input data table as is used in the section Getting Started: CARDINALITY Procedure.

You can load the sashelp.iris data set into your SAS session by using the following DATA step:

data mylib.iris;
   set sashelp.iris;
run;

The following statements run PROC CARDINALITY and plot the levels from the details data set for the variable Species:

proc cardinality data=mylib.iris outcard=mylib.card
                 outdetails=mylib.details maxlevels=5;
run;

data sp;
   set mylib.details( where=(_varname_='Species'));
   label _cfmt_='Formatted value of the variable Species';
   if _index_ = . then do;
      _cfmt_=cats(">",left(_cfmt_));
   end;
  _cfmt_=left(_cfmt_);
run;

proc sgplot data=sp;
  vbar _cfmt_ / freq=_freq_;
run;

Output 7.1.1 shows that the variable Species has only three levels in the details data.

Output 7.1.1: Histogram of Species

Whether MAXLEVELS=100 or 5, the Species variable remains fully visible with a _CARDINALITY_ value of 3.