08-22-2013 12:51 PM
I have 11 sets I'd like to merge into one very wide set, but I want to check for duplicate variables first. For example, I know a number of the sets have a variable named "notes", so I'll need to rename them in each set (notes1, notes2, etc) so they don't get overwritten. Each set has a common ID to merge on.
A PROC COMPARE on two of the sets confirms that "notes" is the only variable I need to worry about, but that just checks two sets. It doesn't seem very efficient to run PROC COMPARE on each combination before the merge (or even 10 combinations off some sort of "master set"). PROC SQL isn't any prettier since it's 10 joins.
I can write a PROC COMPARE macro but that seems like overkill for what's essentially "find duplicate variables among these sets". Does anyone have a systematic way of checking for duplicate variables when merging more than two sets?
Need further help from the community? Please ask a new question.