| NFHS - 4 India Wealth Index [message #15850] |
Fri, 28 September 2018 07:29 |
sankalp.sharma
Messages: 2
Registered: September 2018
|
Member |
|
|
I am currently trying to replicate the wealth index for India provided in the NFHS - 4 household recode. To do so, I was using the official excel file provided on the DHS website ( https://www.dhsprogram.com/what-we-do/survey/survey-display- 355.cfm). However, while cross-checking the summary statistics of all the 132 variables, I noticed that there were 21 variables for which the summary statistics don't match up. I am attaching a list of those variables along with the differences in an excel file.
The larger problem is that due to these discrepancies in the summary statistics, the replicated values of wealth index are not the same as the values provided in the household index. In fact, the correlation between replicated and given values is 0.99973, which is not perfect. Consequently, there are around 9000 observations that end up falling in different quintiles due to these inaccuracies.
Any help/suggestions are appreciated.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
