It can be done easily with SQL subsuery. See below,

data t1;
   do id=1,2,3;
      do j=1,2,3;
             score=round(ranuni(123)*100);
                 if id=2 and j in (2) then score=.;
                 output;
          end;
        end;
        keep id score;
run;

proc print;run;

proc sql;
  select id, score
  from t1
  where id not in (
           select distinct id
           from t1
           where score=.
                   )
  ;
  quit;

assume data=test is such a dataset.

proc sort data=test;
   by ID;
run;

data test_missing (keep=ID temp_index);
   set test;
   if score=.;
   temp_index=1;
run;

data final;
    merge test test_missing;
    by ID;
run;

data final (drop=temp_index);
    set final;
    if temp_index=1 then delete;
run;

/* data=final is what you get */

非常感谢，出来了我想要的结果，谢谢！

非常感谢解答！但是惭愧，我对proc sql 不熟悉，所以就采用3楼那位朋友的做法了。 谢谢！

学习了