Thanks for this. This code makes sense and I ran the merge.
When you say "*this match works fine". What do you mean by that?
I ran the match using the exact code below and only changed the paths. See tab output below.
I understand how some PR (n=54,799) might not have kids, but not how some BR (18,126) might not have households. Only 36% of the data is matching perfectly.
Does that make sense to you? What am I missing?
tab _merge
_merge | Freq. Percent Cum.
------------+-----------------------------------
1 | 18,126 15.89 15.89
2 | 54,799 48.04 63.93
3 | 41,150 36.07 100.00
------------+-----------------------------------
Total | 114,075 100.00
The same for 2007
tab _merge
_merge | Freq. Percent Cum.
------------+-----------------------------------
1 | 9,716 16.75 16.75
2 | 28,459 49.06 65.81
3 | 19,832 34.19 100.00
------------+-----------------------------------
Total | 58,007 100.00
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
