RE: st: RE: Requirement(s) for a match merge

[David Kantor <dkantor@jhu.edu>]

At 05:34 PM 3/30/2005 +0000, Louis wrote:
> Perhaps an example will clarify my concern. Suppose we have the following 
> datasets, where -id- (the identifier) represents family code:
> Dataset1
>      id   father
> 1.   1    John
> 2.   2    Peter
> 3.   3    Sam
>
> Dataset2
>     id  child
> 1.  1   Bill
> 2.  2   Joe
> 3.  2   Mary
> 4.  3   Phil
> 5.  3   Sue
> 6.  3   Pat
>
> I was thinking that even though -id- does not uniquely identify each 
> observation in dataset 2, the datasets can still be merged using -id- as 
> the match variable to obtain a result like the following:
>
>     id   father child
> 1.  1    John   Bill
> 2.  2    Peter  Joe
> 3.  2    Peter  Mary
> 4.  3    Sam    Phil
> 5.  3    Sam    Sue
> 6.  3    Sam    Pat
>
> Or is it that this kind of merge is not called match merge? Thanks.
>
> Louis
> [...]
That's exactly what I meant in my last message -- that the identifier 
uniquely identifies observations in one file but not the other.  I do call 
this a match merge, and I believe that StataCorp does as well.
--David

David Kantor
Institute for Policy Studies
Johns Hopkins University
dkantor@jhu.edu
410-516-5404
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/