Combine 2 sas data sets (no matching criteria)

[GreggB]

SAS data set 1 columns:     Code   Grade  DocType  SecNum

SAS data set 2 columns:     Last First ID  form

Data set I want:                   Code   Grade  DocType  SecNum  Last First ID  form

Basically, a cut/paste if these were Excel files...but I keep getting the Cartesian product.

[Reeza]

No BY statement with a data step merge?

[GreggB]

no merge. 

[Reeza · Accepted solution]

---

@GreggB wrote:

no merge. 

---

Why not?

data this;
var1 = 'This1';
var2 = 'That1';
run;
data that;
var3 = 'This2';
var4 = 'That2';
run;

data want1;
merge this that;
run;

data want2;
set this that;
run;

proc print data=want1;
proc print data=want2;
run;

Results:

via MERGE

Obs	var1	var2	var3	var4
1	This1	That1	This2	That2

---

Via SET

Obs	var1	var2	var3	var4
1	This1	That1
2			This2	That2

[GreggB]

this works! I always thought a MERGE had to have a BY statement

[Reeza]

---

@GreggB wrote:

this works! I always thought a MERGE had to have a BY statement

---

Nope, that's what my first answer attempted to say, but it was worded badly.

[VDD]

@GreggB you have no variable in both of the datasets so merging will not work.

you results are dataset1 and appended dataset2.

you have to have something in common in both datasets to merge and retrieve your wants.

data this;
var1 = 'This';
var2 = 'That';
run;
data that;
var3 = 'This';
var4 = 'That';
run;

data results;
set this that;
run;

[GreggB]

I don't want to merge them. I want columns A, B and C in set 1 and columns D, E and F in set 2 to be put in a new data set that has columns A, B, C, D, E, F.

[Astounding]

This is a possibility:

data want;
   set dataset1;
   set dataset2;
run;

However, it's not clear how  many observations you are working with in each data set.  Would it be possible that one data set has 3 observations, and the other has 5 observations?  If that's the case, how would you like them to be matched up?

[GreggB]

I don't want them matched. Maybe "overlay" is the right word. 

[Astounding]

You should actually try the program I suggested.  If you have one observation in each data set, it does exactly what you are asking for.

[VDD]

@GreggB which part of @Reeza answered your needs the merge or datastep?

[GreggB]

merge