Copy first value in variable to all observations for each id

[Tom]

Yes. It is only use the POINT= on the first observation so KEEP= is not needed as values re-read will be the same values.

The easiest to understand method is to use the RETAIN statement to keep the value of SCORE from the first observation.

data want;
  set reading;
  by id score;
  if first.score then low_score=score;
  retain low_score;
run;

Two things to check.

1) The variable retained (LOW_SCORE) cannot exist on the source dataset (READING) or else its value will be overwritten when the SET statement next executes.

2) If there are missing values of SCORE they will sort before any valid value.  In that case you might find the PROC SQL solution easiest to understand.

proc sql;
  create table want as
    select *,min(score) as low_score
    from reading
    group by id 
  ;
quit;

Note this uses a SAS specific feature where it will re-merge the LOW_SCORE value calculated at the ID level back onto all of the individual observations for that value of ID.

Or possibly the double "DOW" (do while) loops.

data want;
do until(last.id);
  set reading;
  by id;
  low_score=min(low_score,score);
end;
do until(last.id);
  set reading;
  by id;
  output;
end;
run;