Use PROC STDIZE. This will subtract the mean from every observation, then you can count the number that are positive (above average).
proc stdize data=tabelle method=mean out=tabelle1;
var tunrover;
run;
proc format;
value highlow low-<0='Below Avg' 0='Avg' 0<-high='Above Avg';
run;
proc freq data=tabelle1;
table tunrover;
format tunrover highlow.;
run;
For your future benefit, saying something didn't work and not showing us the code you used, is not helpful. For us to help you, we need to see the code that didn't work. (We don't need to see the code that doesn't work here, only because I have shown code that uses a different approach)
Another comment: don't try to write your own DATA step code that does something that SAS has already included in a PROC. Why? Because the PROC has been thoroughly tested and debugged by SAS, so you don't have to do those things. You know the PROC works properly, whereas if you write your own DATA step code, you can get it wrong, as you have seen in this case.
--
Paige Miller