| Re: De-normalizing weights and svyset command in Stata [message #5558 is a reply to message #3620] |
Sat, 06 June 2015 07:33   |
Mercysh
Messages: 35
Registered: April 2014
|
Member |
|
|
I am sorry I lost you in the 3rd paragraph, where Tom says that you can not rely on V000 because it is a string variable. What does one need to do if using two DHS waves/phases for three countries for example to have a unique identifier for "surveys" for each country and wave (one survey per phase). Secondly would a change in the regions need to be considered when pooling the data e.g. 2004 Malawi DHS was representative at selected districts (in addition to national) and the 2010 at all districts except that two were combined into one for survey purposes. I want to use svy prefix in Stata. Your help will be highly appreciated.
Mercy
[Updated on: Sat, 06 June 2015 07:55] Report message to a moderator |
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
