DATA Step, Macro, Functions and more

Merging two large datasets with SAME variable names

Reply
New Contributor
Posts: 2

Merging two large datasets with SAME variable names

Hello,

I am merging two rather large datasets each of which contain 4,000 variables and they both have the same variable names.

Given the large number of variables, I am wondering if there a SAS procedure that would allow me to rename all of the variables in a single dataset at once? Perhaps by adding a simple suffix or prefix?

 

I Googled around and found some PROC SQL options however, I am simply not familiar  with SQL so the code was not very clear.

 

I am using SAS version 9. 

 

Any help or references would be appreciated.

 

Thank you.

Super User
Posts: 19,780

Re: Merging two large datasets with SAME variable names

If they have the same variable names are you sure you want a merge and not an append?

 

Although it doesn't matter for this question you a SAS version is along the lines of 9.3 or 9.4 not just 9. 

New Contributor
Posts: 2

Re: Merging two large datasets with SAME variable names

OH!!! Yes... well I think I want to Append.
One dataset includes the children and the other include the mothers they
are to be merged(?) using a family ID.
Both groups completed the same questionnaire (hence the same variable
names) however now I am tasked with bringing them both together.
Appending sounds like a reasonable direction. I will Google that
procedure, right now!

##- Please type your reply above this line. Simple formatting, no
attachments. -##
Super User
Posts: 7,771

Re: Merging two large datasets with SAME variable names

Well, that's quite simple, then.

data family;
set
  mothers
  children
;
run;

proc sort data=family;
by family_id;
run;
---------------------------------------------------------------------------------------------
Maxims of Maximally Efficient SAS Programmers
Super User
Super User
Posts: 7,942

Re: Merging two large datasets with SAME variable names

4000 variables in multiple datasets, that really does not sound like an efficient or easy way to be working.  Personally I wouldn't work with any dataset having more than 30 odd variables, its just unworkable.  Consider a restructure.

Ask a Question
Discussion stats
  • 4 replies
  • 506 views
  • 0 likes
  • 4 in conversation