| Relative index of inequality calculation [message #11195] |
Thu, 17 November 2016 02:16  |
zobidaislam
Messages: 6
Registered: March 2016
Location: Dhaka
|
Member |
|
|
Hi,
I am using Pakistan DHS data (women's data). I want to estimate the relative index of inequality for antenatal care (Inequality based on wealth index). If I just use logistic regression model without adjusting for other factors, can I call it "relative index of inequality"?
If my dependent variable is anc (0=no, 1=yes) and the covariate is quintile (1=poorest, 2=poorer, 3=middle class, 4=richer, 5= richet)
The code is:
char quintile [omit] 1
logistic anc i.quintile
Thank you.
Zobi
|
|
|
|
| Re: Relative index of inequality calculation [message #11198 is a reply to message #11195] |
Thu, 17 November 2016 10:45  |
Bridgette-DHS
Messages: 3035
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
Quote:
I assume that you are looking at Comparative Report 42, page 5, index 2. What you have described is correct, but be sure to use svyset and svy to adjust for the weights, clusters, and strata. You want the coefficient for the fifth quintile, without exponentiation.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
