Removing duplicate observations with missing fields

[Lietuvis19]

Hi All,

I am trying to remove some duplicated observations from my data set. The goal is to end with a unique student id and any course they are affiliated with.

For the duplicate observations (same student id and the same course)

• if there is a value in the grade variable it should be kept
• if there is no value in the grade variable it should be represented as missing

This is my first post on here, sorry if I missed anything!

This is how my data looks like right now:

studentid	Grade	final_grade	course
1		A	MATH
1		A	MATH
1	C	B	ECON
1	B	A	ENG
1	B	A	ENG
1		C	SCI
1	F	C	SCI
1		B	CHEM
2	D	B	ENG
2		B	ENG
2		A	CHEM
3	F	C	MATH

This is what my end goal should look like. Each student having only 1 observation for whatever course they have taken. 

studentid	Grade	final_grade	course
1		A	MATH
1	C	B	ECON
1	B	A	ENG
1	F	C	SCI
1		B	CHEM
2	D	B	ENG
2		A	CHEM
3	F	C	MATH

I feel like this is a simple answer, but I have scavenged the web with no use. I was originally thinking of copying the grade into the missing fields and simply deleting the duplicate observations? Started running into issues with identifying unique student id and unique course for those students.

Thank you very much!

[novinosrin · Accepted solution]

data have;
input studentid	Grade$	final_grade	$ course $;
cards;
1	 .	A	MATH
1	 .	A	MATH
1	C	B	ECON
1	B	A	ENG
1	B	A	ENG
1	. 	C	SCI
1	F	C	SCI
1	 .	B	CHEM
2	D	B	ENG
2	. 	B	ENG
2	 .	A	CHEM
3	F	C	MATH
;


proc sort data=have out=_have;
by studentid course;
run;
data want;
update _have(obs=0) _have;
by studentid course ;
run;