Merging DHS data sets with municipality information [message #14330] |
Fri, 23 March 2018 05:15 |
clotilde.mahe
Messages: 3 Registered: March 2018
|
Member |
|
|
Dear all,
I would like to merge municipality information for a different source with the different DHS data sets (for 1995, 2000, 2005 and 2010 DHS waves for Colombia). However, I am having difficulties understanding which variable(s) give the municipality number in the DHS data sets. 'smunip' is a code from 1 to 61, specific to the DHS for surveyed municipalities that cannot be used with the official (DANE) used to code municipalities... The variable giving information on the primary sampling area (a municipality according to the report of 1995) ('smuest', if I am not mistaken) contains much more unique observations than the sampled number of municipalities (61)... Is there a way to obtain the DANE code of municipalities in the DHS data sets so that I could merge information on municipalities from another source?
Thank you for your consideration and help.
Clotilde
[Updated on: Fri, 23 March 2018 05:15] Report message to a moderator
|
|
|
|
|
|
|
|
|
|
Re: Merging DHS data sets with municipality information [message #26313 is a reply to message #24070] |
Tue, 07 March 2023 09:46 |
manhar
Messages: 2 Registered: March 2023
|
Member |
|
|
Hi!
I am still unable to get the DANE municipality codes for DHS 1995 or 2000 using the instructions provided (using shmunip and shdepto). I understand this is an old thread, but any idea if it is possible to get these from these DHS rounds for Colombia? For both 1995 and 2000, the shmunip variable takes on values from 1 to 61- which, after concatenating with the department will not yield the desired DANE municipality codes.
I would hugely appreciate the help!
Thanks and regards,
Manhar
[Updated on: Tue, 07 March 2023 11:04] Report message to a moderator
|
|
|
Re: Merging DHS data sets with municipality information [message #26317 is a reply to message #26313] |
Tue, 07 March 2023 13:02 |
Bridgette-DHS
Messages: 3189 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
Guillermo Rojas first told you on March 23, 2018, almost 5 years ago, that in the 2015 survey the code is given by code=1000*shdepto + shmunip; the code is calculated similarly in all other surveys except 1995; and the information to calculate the code in the 1995 survey is unfortunately absent. Variable names such as "shdepto" and "shmunip", beginning with "sh", are in the HR and PR files. The variable names may begin with other prefixes in other files and the variable names may be different in other surveys. We (DHS staff) have nothing to add to this. Perhaps some other users can help.
|
|
|