| Indonesia DHS 2012 urban/rural by census block and village [message #5511] |
Tue, 02 June 2015 08:46  |
amyfinnegan
Messages: 5
Registered: March 2013
Location: Durham, NC
|
Member |
|
|
Hello -
I am working with the 2012 Indonesia DHS. I am linking the DHS to the PODES (Village Potential Surveys) in Indonesia by the village codes. I have created these 10 digit codes by concatenating the DHS variables v024, sregmun, ssubd and svillag. First, I have inserted leading zeros so that sregmun is 2 digits, ssubd is 3 digits and svillag is 3 digits. When I tabulate the village codes by v025 (urban/rural), I find that some observations (sometimes 1 or 2) within a village may be classified as urban with the rest (~25) being classified as rural or vice versa. I find the same thing when I tabulate v001 (cluster number/PSU) with urban/rural. I would have thought that census blocks (the PSU in this survey) and villages should be either urban or rural but not both. Additionally, I find that I have more villages than census blocks (1,943 vs. 1,832). Lastly, I also find that when I give each PSU-househould number a unique id by concatenating v001 and v002 that some households are classified as both urban and rural.
Any help sorting this out would be much appreciated!
Many thanks,
Amy
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
