Preventing overwrite

[learn2]

Hi I have two datasets that I would like to merge by id, but when I do that, I have Dataset B overwriting the diagnosis date, cluster and servdate in Dataset A, how do I prevent this?

Dataset A:

ID	Diagnosis Date	Cluster	Servdate
1	Jan 1 1998	Urban	Aug 3 1998
1	Jan 1 1998	Urban	Aug 15 1998
2	Feb 3 2000	Urban	April 9 2000

Dataset B:

ID	Diagnosis Date	Cluster	Servdate
1	Jan 1 1998	Rural	Sept 1 1998
1	Jan 1 1998	Rural	Sept 3 1998
2	Feb 3 2000	Rural	May 7 2000

[novinosrin]

Decide which ones you want to keep and drop the rest using dataset options

[learn2]

This is what I need:

ID	Diagnosis Date	Cluster	Servdate
1	Jan 1 1998	Urban	Aug 3 1998
1	Jan 1 1998	Urban	Aug 15 1998
1	Jan 1 1998	Rural	Sept 1 1998
1	Jan 1 1998	Rural	Sept 3 1998
2	Feb 3 2000	Urban	April 9 2000
2	Feb 3 2000	Rural	May 7 2000

[Astounding]

What would you like the result to be?

Seeing what you posted as the result, change the MERGE statement to a SET statement.

[learn2]

Thank you!

[novinosrin]

data want;

set a b;

by id;

run

What you need is interleave and not a merge

[novinosrin · Accepted solution]

worked out example using your sample

data a;
infile cards truncover;
input (ID	DiagnosisDate	Cluster	Servdate) (& $15.);
cards;
1	Jan 1 1998	Urban	Aug 3 1998
1	Jan 1 1998	Urban	Aug 15 1998
2	Feb 3 2000	Urban	April 9 2000
;

data b;
infile cards truncover;
input (ID	DiagnosisDate	Cluster	Servdate) (& $15.);
cards;
1	Jan 1 1998	Rural	Sept 1 1998
1	Jan 1 1998	Rural	Sept 3 1998
2	Feb 3 2000	Rural	May 7 2000
;

data want;
set a b;
by id;
run;

[novinosrin]

Or if your datasets are not sorted by id , you could use proc sql

proc sql;
create table want as
select * from a
union all
select * from b
order by id, DiagnosisDate,Servdate;
quit;