This was perfect! Thank you ... View more

 What if there is a tie? If there is a tie I need to check another database  Your sample seems to assume that SSN 111111 (first row) is the same person as SSN 1111111111 rows 2 through 4).  Is that just a typo? Yes this was a typo - it should be all the same SSN for the example Is your data already sorted by ssn? Yes Do you want to reduce multiple records to one as in your example?  This would be a problem if your original records have variables besides SEX that vary over a set of matching SSN's. -- I need to check multiple variables that have the same issue (take the most recurring value for each variable I need). My idea was to do this in segments to make it less complicated and then create a new dataset with one line per person with all of the necessary demographic information  ... View more