Wealth quantiles [message #26628] |
Tue, 11 April 2023 11:58 |
chande
Messages: 12 Registered: March 2021
|
Member |
|
|
Dear users,
I hope all is well
I am working on African countries' DHS. In the Madagascar HH file and IR file, I focused on the v190 variable to see the wealth quantiles. I wanted to know how these quantiles are not equal in proportion (20%). As in the dataset, the proportion is different and the higher share is for the richest quantile (around 24%) compared to the poorest one.
If i make the quantiles with the wealth index score then its makes equal quantiles. I wanted to know why v190 has different quantiles as one of my main variables in the analysis.
|
|
|
Re: Wealth quantiles [message #26639 is a reply to message #26628] |
Wed, 12 April 2023 08:51 |
Janet-DHS
Messages: 855 Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS staff member, Tom Pullum:
The wealth quintiles are constructed using household-level data. Everyone in the same household is in the same quintile (hv270). In terms of the number of de jure household members in the entire survey, the quintiles are equal in size.
When the other files are constructed, hv270 is carried into the other files but with a different name--v190 in the IR, BR, and KR files and mv190 in the MR (men's) file. In these files, for most countries, the number of individuals is not the same in every quintile. The difference is usually the greatest in the KR file, because of the typical inverse relationship between wealth and fertility. Households in the lowest wealth quintile tend to have (that is, the women in these households tend to have) more children.
If you reconstruct the boundaries for the quintiles you will lose comparability with all the other analyses that have not done that, so I advise against such a change.
|
|
|