mbliss,
If the updating field is also residing in a separate SAS DATASET then you would merge two dataset, dropping the original value and retaining new one.
/**dataset to be updated
drop the original field1 sort by the merge field here it is "key"**/
proc sort data=dsn.data1 out=data1(drop=field1);
by key;
run;
/***dataset that has new values that would be used to do update**/
proc sort data=dsn.data2 out=data2(keep=key field1);
by key;
run;
data new;
merge
data1(in=in1)
data2(in=in2)
; by key;
if in1 and in2;/**this is an inner join**/
run;