The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in Stata » merging WI with IR/MR for Zambia (code for South Africa will not work)
merging WI with IR/MR for Zambia (code for South Africa will not work) [message #1121] Thu, 09 January 2014 09:32 Go to next message
alitasso is currently offline  alitasso
Messages: 5
Registered: January 2014
Member
I am trying to merge the WI file with the IR/MR files for Zambia 2001-02. In this post (in case the URL dies, the post is titled, "merging wealth index from Ethiopia DHS 2000 to women's file"), Tom Pullum provides some code to merge the WI file with the IR file for Ethiopia 2000. However, it does not seem to work for Zambia.

First I look at the range of v001 and v002 values in the IR file:

. use ZMIR42FL.dta, clear

. tabstat v001 v002, stat(min max)

stats | v001 v002
---------+--------------------
min | 1 1001
max | 320 9469
------------------------------

Then I try to apply Tom Pullum's code to the WI file:

. use ZMWI41FL.dta, clear

. gen str9 wv001=substr(whhid,1,9)

. gen str9 wv002=substr(whhid,10,3)

. destring wv001, gen(v001)
wv001 has all characters numeric; v001 generated as int

. destring wv002, gen(v002)
wv002 has all characters numeric; v002 generated as int

. tabstat v001 v002, stat(min max)

stats | v001 v002
---------+--------------------
min | 15 1
max | 3201 469
------------------------------

As you can see, the range of v001 and v002 codes generated by Tom Pullum's code does not match the range of v001 and v002 codes that are in the IR data. It looks like it's just off by just one column: v001 should be 3 digits, and v002 should be 4 digits. But it doesn't seem to be that simple.

If I modify the code slightly, then I get some unwanted spaces in v002:

. use ZMWI41FL.dta, clear

. gen str9 wv001=substr(whhid,1,8)

. gen str9 wv002=substr(whhid,9,4)

. destring wv001, gen(v001)
wv001 has all characters numeric; v001 generated as int

. destring wv002, gen(v002)
wv002_ contains nonnumeric characters; no generate

However, this code seems to work fine for v001:

. tabstat v001, stat(min max)

variable | min max
-------------+--------------------
v001 | 1 320
----------------------------------

I tried a crude solution of just deleting the spaces in wv002, but then I get a variable that does not range from 1 to 9469 (as it should):

. gen str9 wv002_=subinstr(wv002," ","",.)

. destring wv002_, gen(v002_)
wv002_ has all characters numeric; v002_ generated as int

. tabstat v002_, stat(min max)

variable | min max
-------------+--------------------
v002_ | 11 9469
----------------------------------

Any assistance would be appreciated -- thank you!

[Updated on: Sun, 02 February 2014 08:16]

Report message to a moderator

Re: merging WI with IR/MR for Zambia (code for South Africa will not work) [message #1439 is a reply to message #1121] Wed, 26 February 2014 09:09 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member
We are working on a response to your posting. Thanks!
Re: merging WI with IR/MR for Zambia (code for South Africa will not work) [message #1444 is a reply to message #1121] Wed, 26 February 2014 10:27 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member
Attached is a text file from DHS Specialist, Tom Pullum. He suggests that you modify the paths in the attachment and then cut/paste the syntax into the command window of Stata.
Previous Topic: I need help identifying families who live in the same dwelling.
Next Topic: DHS 2012 Ecuador
Goto Forum:
  


Current Time: Thu Mar 28 05:38:25 Coordinated Universal Time 2024