label duplicate cases in data set

[fengyuwuzu]

My data set has variables user_id and others like v1, v2, .. v5.

There are some duplicate cases which have the same user_id.

is there a way to label the duplicate cases, by adding a new variable, value 1 for the primary cases, and value 0 for the second case with the same user_id? By doing this I do not need to delete the duplicate cases, but when analysis I can select the distinct cases by this new variable.

Thanks in advance.

[Astounding · Accepted solution]

Sort your data if it's not yet in order:

proc sort data=have;

by user_id;

run;

Then it's simple:

data want;

set have;

by user_id;

new_variable = first.user_id;

run;

[fengyuwuzu]

great! this works!

This forum is really good.