| V001 and V002 do not uniquely identify observations in certain files? [message #931] |
Mon, 18 November 2013 21:27  |
cp284
Messages: 3
Registered: November 2013
|
Member |
|
|
Hi everyone,
I'm trying to merge individual and household files for a number of countries. Per the DHS documentation on merging, I've renamed the variables HV001 and HV002 to V001 and V002 in the household files, and then merge on these two variables. This has worked for almost all of my files of interest. But for 12 household files, V001 and V002 do not uniquely identify observations. This prevents me from carrying out the m:1 merge in Stata that I have done for my other files.
V001 and V002 do not uniquely identify observations in the following files:
BOHR31FL
CMHR31FL
TFHR31FL
NIHR31FL
TGHR31FL
MLHR41FL
BJHR41FL
BOHR41FL
ETHR41FL
SNHR44FL
IAHR42FL
HNHR52FL
Has anyone else come across this problem before? How have you navigated it?
Thank you!
|
|
|
|
| Re: V001 and V002 do not uniquely identify observations in certain files? [message #952 is a reply to message #931] |
Mon, 02 December 2013 09:30  |
Bridgette-DHS
Messages: 3230
Registered: February 2013
|
Senior Member |
|
|
Here is a response from one of our Senior Data Processing Specialists, Noureddine Abderrahim.
In some surveys HV002, V002, or MV002 are not sufficient to identify the households. In some survey a household is identified by the household number and the structure within this household. In such a case, you might use the first 12 characters of the "CASEID" identification of the woman or the "MCASEID" identification of the man to match them with their respective household identified by the variable "HHID".
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
