I have a dataset of 60K rows and I need to remove duplicates that meet certain criteria across multiple variables.   The values flagged and removed have to meet all three of the criteria as follows: Criteria 1: Duplicates of variable 1  Criteria 2. That contains 2 or more of the same value in variable 2 Criteria 3. That contains the same value (1) across ALL rows (that meets the first and second criteria) in variable 3.    Below is an example of the data I have, the data I want, and the data I want to exclude.    Example data I have: Variable1 Variable2 Variable3 55 1 1 55 1 1 55 1 1 21 1 1 21 2 2 21 2 1 33 1 2 90 2 1 90 3 1 90 2 1 67 2 1 67 2 1 67 2 1 67 1 1 67 1 1 67 1 1 40 3 1 81 6 2 81 6 1 81 4 1 81 4 1 43 2 2 43 2 1 21 9 1 21 9 1 21 9 1 21 9 1 55 2 1 55 2 1   Example data of what I want to keep: Variable1 Variable2  Variable3 21 1 1 21 2 2 21 2 1 33 1 2 40 3 1 43 2 2 43 2 1   Example data that I need to flag then filter out: Variable1 Variable2  Variable3 55 1 1 55 1 1 55 1 1 90 2 1 90 2 1 67 2 1 67 2 1 67 2 1 67 1 1 67 1 1 67 1 1 81 4 1 81 4 1 21 9 1 21 9 1 21 9 1 21 9 1 55 2 1 55 2 1     I am new to SAS programming so please forgive my naivete. I searched the forums and found the code below but it did not work, I receive syntax errors. Perhaps I am putting in the variable names incorrectly. Either way, I am not even sure if this code is what I need.  I appreciate your help...   PROC SQL; CREATE TABLE WORK.Orders1 AS SELECT t1.variableone, t1.variabletwo, t1.variablethree FROM WORK.Orders t1 ORDER BY t1.variableone, t1.variabletwo, t1.variablethree; QUIT; data WORK.Orders2; set WORK.Orders1; by FIRST.WORK.Orders1.variableone LAST.WORK.Orders1.variabletwo ; if not (WORK.orders1.one and WORK.orders1.two) then flag_1=1; else flag_1=0; run; ... View more