The identification of the input data set is performed by checking the IN= variables defined for each input data set in the SET statement.

data combined;
  set sashelp.prdsale(in=in_prdsale)
      sashelp.prdsal2(in=in_prdsal2)
      sashelp.prdsal3(in=in_prdsal3);
  length Source $ 32;
  if in_prdsale then source="PRDSALE";
  else if in_prdsal2 then source="PRDSAL2";
  else if in_prdsal3 then source="PRDSAL3";
run;
proc freq data=combined;
  tables Source;
  title1 "Source Data Sets for Combined Data";
run;

SAS Log

6991   data combined;
6992     set sashelp.prdsale(in=in_prdsale)
6993         sashelp.prdsal2(in=in_prdsal2)
6994         sashelp.prdsal3(in=in_prdsal3);
6995     length Source $ 32;
6996     if in_prdsale then source="PRDSALE";
6997     else if in_prdsal2 then source="PRDSAL2";
6998     else if in_prdsal3 then source="PRDSAL3";
6999   run;
 
NOTE: There were 1440 observations read from the data set SASHELP.PRDSALE.
NOTE: There were 23040 observations read from the data set SASHELP.PRDSAL2.
NOTE: There were 11520 observations read from the data set SASHELP.PRDSAL3.
NOTE: The data set WORK.COMBINED has 36000 observations and 15 variables.
 
7000   proc freq data=combined;
7001     tables Source;
7002     title1 "Source Data Sets for Combined Data";
7003   run;
 
NOTE: There were 36000 observations read from the data set WORK.COMBINED.
 

SAS Listing Output

Source Data Sets for Combined Data
 
The FREQ Procedure
 
                                    Cumulative    Cumulative
Source     Frequency     Percent     Frequency      Percent
------------------------------------------------------------
PRDSAL2       23040       64.00         23040        64.00
PRDSAL3       11520       32.00         34560        96.00
PRDSALE        1440        4.00         36000       100.00

An alternate approach is to use the INDSNAME= option in the SET statement, available starting in SAS 9.2.

The variable defined by the INDSNAME= option contains the complete input data set including the libref, for a maximum length of 41 characters.

This technique is particularly valuable when combined with the data set list syntax introduced in SAS 9.2.

data combined;
  set sashelp.prdsal: indsname=DataSetName;
  length Source $ 32;
  Source=scan(DataSetName,2);
run;
proc freq data=combined;
  tables Source;
  title1 "Source Data Sets for Combined Data";
run;

SAS Log

7015   data combined;
7016     set sashelp.prdsal: indsname=DataSetName;
7017     length Source $ 32;
7018     Source=scan(DataSetName,2);
7019   run;
 
NOTE: There were 23040 observations read from the data set SASHELP.PRDSAL2.
NOTE: There were 11520 observations read from the data set SASHELP.PRDSAL3.
NOTE: There were 1440 observations read from the data set SASHELP.PRDSALE.
NOTE: The data set WORK.COMBINED has 36000 observations and 15 variables.
 
7020   proc freq data=combined;
7021     tables Source;
7022     title1 "Source Data Sets for Combined Data";
7023   run;
 
NOTE: There were 36000 observations read from the data set WORK.COMBINED.
 

SAS Listing Output

Same as for the old way

Advantages of the alternate approach:

• No data set options are required for the input data sets.
• No conditional logic is required to create the input source variable.

Disadvantages of the alternate approach:

• The value of the input data set obtained through the INDSNAME= option is always uppercase.