The original approach is to collapse each variable down to unique values, then count the number of collapsed items. The counts for individual variables subsequently are concatenated to create a single result set.

data Georgia;
  set sashelp.zipcode(where=(StateName="Georgia"));
  if msa=0 then msa=.;
run;
proc freq data=georgia;
  tables CountyNm / out=Counties noprint;
  tables AreaCode / out=AreaCodes noprint;
  tables ZIP_Class / out=ZIPclasses noprint;
  tables MSA / out=MSAs noprint;
run;
data counts;
  set Counties(in=county_level)
      AreaCodes(in=areacode_level)
      ZIPclasses(in=zipclass_level)
      MSAs(in=msa_level)
        end=no_more;
  if county_level then CountyCount+1;
  else if areacode_level then AreaCodeCount+1;
  else if zipclass_level then ZIPclassCount+1;
  else if msa_level then MSACount+1;
  if no_more then output;
  keep CountyCount AreaCodeCount ZIPclassCount MSACount;
run;
proc print data=counts noobs;
  title1 "Counts of Variable Levels";
run;

SAS Log

6798   data Georgia;
6799     set sashelp.zipcode(where=(StateName="Georgia"));
6800     if msa=0 then msa=.;
6801   run;
 
NOTE: There were 972 observations read from the data set SASHELP.ZIPCODE.
      WHERE StateName='Georgia';
NOTE: The data set WORK.GEORGIA has 972 observations and 18 variables.
 
6802   proc freq data=georgia;
6803     tables CountyNm / out=Counties noprint;
6804     tables AreaCode / out=AreaCodes noprint;
6805     tables ZIP_Class / out=ZIPclasses noprint;
6806     tables MSA / out=MSAs noprint;
6807   run;
 
NOTE: There were 972 observations read from the data set WORK.GEORGIA.
NOTE: The data set WORK.COUNTIES has 159 observations and 3 variables.
NOTE: The data set WORK.AREACODES has 6 observations and 3 variables.
NOTE: The data set WORK.ZIPCLASSES has 3 observations and 3 variables.
NOTE: The data set WORK.MSAS has 9 observations and 3 variables.
 
6808   data counts;
6809     set Counties(in=county_level)
6810         AreaCodes(in=areacode_level)
6811         ZIPclasses(in=zipclass_level)
6812         MSAs(in=msa_level)
6813           end=no_more;
6814     if county_level then CountyCount+1;
6815     else if areacode_level then AreaCodeCount+1;
6816     else if zipclass_level then ZIPclassCount+1;
6817     else if msa_level then MSACount+1;
6818     if no_more then output;
6819     keep CountyCount AreaCodeCount ZIPclassCount MSACount;
6820   run;
 
NOTE: There were 159 observations read from the data set WORK.COUNTIES.
NOTE: There were 6 observations read from the data set WORK.AREACODES.
NOTE: There were 3 observations read from the data set WORK.ZIPCLASSES.
NOTE: There were 9 observations read from the data set WORK.MSAS.
NOTE: The data set WORK.COUNTS has 1 observations and 4 variables.
 
6821   proc print data=counts noobs;
6822     title1 "Counts of Variable Levels";
6823   run;
 
NOTE: There were 1 observations read from the data set WORK.COUNTS.
 

SAS Listing Output

Counts of Variable Levels
 
           Area
County     Code    ZIPclass
 Count    Count      Count     MSACount
 
  159       6          3           9

An alternate approach is to apply the NLEVELS option in PROC FREQ, using the ODS OUTPUT statement to capture the counts into a SAS data set.

data Georgia;
  set sashelp.zipcode(where=(StateName="Georgia"));
  if msa=0 then msa=.;
run;
ods output NLevels=CountsNLevels;
proc freq data=georgia nlevels;
  tables CountyNm AreaCode ZIP_Class MSA  / noprint missing;
  label ZIP_Class=" " MSA=" " CountyNm=" " AreaCode=" ";
  title1 "Counts of Variable Levels";
run;
proc print data=CountsNLevels;
  title1 "Counts of Variable Levels";
run;

SAS Log

6835   data Georgia;
6836     set sashelp.zipcode(where=(StateName="Georgia"));
6837     if msa=0 then msa=.;
6838   run;
 
NOTE: There were 972 observations read from the data set SASHELP.ZIPCODE.
      WHERE StateName='Georgia';
NOTE: The data set WORK.GEORGIA has 972 observations and 18 variables.
 
6839   ods output NLevels=CountsNLevels;
6840   proc freq data=georgia nlevels;
6841     tables CountyNm AreaCode ZIP_Class MSA
6842    / noprint missing;
6843     label ZIP_Class=" " MSA=" " CountyNm=" " AreaCode=" ";
6844     title1 "Counts of Variable Levels";
6845   run;
 
NOTE: The data set WORK.COUNTSNLEVELS has 4 observations and 4 variables.
NOTE: There were 972 observations read from the data set WORK.GEORGIA.
 
6846   proc print data=CountsNLevels;
6847     title1 "Counts of Variable Levels";
6848   run;
 
NOTE: There were 4 observations read from the data set WORK.COUNTSNLEVELS.
 

SAS Listing Output

Counts of Variable Levels
 
The FREQ Procedure
 
           Number of Variable Levels
 
                          Missing    Nonmissing
Variable       Levels      Levels        Levels
-----------------------------------------------
COUNTYNM          159           0           159
AREACODE            6           0             6
ZIP_CLASS           3           1             2
MSA                 9           1             8
========================================================================================
Counts of Variable Levels
 
                                                NNon
                                   NMiss        Miss
Obs    TableVar      NLevels      Levels      Levels
 
 1     COUNTYNM          159           0         159
 2     AREACODE            6           0           6
 3     ZIP_CLASS           3           1           2
 4     MSA                 9           1           8

Advantages of the alternate approach:

• A single PROC step can analyze any number of variables.
• In addition to determining the number of unique values, levels of missing values are identified.
• If only a report is required, there is no need to create a data set to hold the results.

Disadvantages of the alternate approach:

• This approach produces a multiple-observation result set that must be transposed if one column per original variable is desired.