Hi,  I do not know if this what you are looking for (based on the istructions in the notebook - picture, it is counting status M) :   DATA TEST; INPUT MARITAL_STATUS $ INITIAL_BP FINAL_BP; DATALINES; M 60 70 S 80 60 C 90 70 C 100 60 M 120 80 S 141 90 S 80 120 S 90 110 S 115 110 M 117 112 C 114 112 C 105 96 M 103 87 M 107 110 S 106 110 ;   PROC SQL ; SELECT COUNT(*) INTO :NS FROM TEST WHERE MARITAL_STATUS='S'; SELECT COUNT(*) INTO :NNS FROM TEST WHERE MARITAL_STATUS='S' & SUM(INITIAL_BP,-FINAL_BP)<0 ; QUIT;   %PUT SINGLES &NS. ;  /*can be omitted*/ %PUT SINGLES NEGATIVE &NNS. ; /*can be omitted*/ DATA TEST; SET TEST;      IF SUM(INITIAL_BP,-FINAL_BP)<0 & MARITAL_STATUS='S' THEN                    JIF=(INITIAL_BP**2/SUM(INITIAL_BP,-FINAL_BP)**3)/(&NS.*&NNS.); RUN;   PROC PRINT DATA=TEST; RUN; PROC MEANS DATA=TEST N SUM;       VAR JIF; RUN;       Obs MARITAL_STATUS INITIAL_BP FINAL_BP JIF 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 M 60 70 . S 80 60 . C 90 70 . C 100 60 . M 120 80 . S 141 90 . S 80 120 -0.00556 S 90 110 -0.05625 S 115 110 . M 117 112 . C 114 112 . C 105 96 . M 103 87 . M 107 110 . S 106 110 -9.75347       Analysis Variable : JIF N Sum 3 -9.8152778 ... View more