| Multiple variable names [message #3163] |
Tue, 28 October 2014 09:05  |
DHS user
Messages: 111
Registered: February 2013
|
Senior Member |
|
|
I want to ask about the birth weight variable or m19 which contain m19_1 m19_2 etc. What is the difference among them. I wonder if I can use only the first or I have to merge those variables.
|
|
|
|
| Re: Multiple variable names [message #3164 is a reply to message #3163] |
Tue, 28 October 2014 09:08  |
Bridgette-DHS
Messages: 3035
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Specialist, Tom Pullum:
You are apparently working with the IR file, in which the woman is the unit of analysis. The suffixes _1, _2, _3, _4, _5, _6, refer to her children born in the past five years. _1 refers to the most recent birth in the past five years, _2 to the next most recent, etc. If the woman did not have any children in the past five years, then the variables with these suffixes in the IR file will all be given a missing value (not applicable) code, which in Stata is a period, ".". If she only had one birth in the past five years, then only variables with the suffix _1 will be non-missing, etc.
To analyze such variables we normally use another file that has children born in the past five years as the units of analysis. This is the KR file. It includes m19 but without suffixes. Each record in this file includes virtually all of the woman's (mother's) data.
|
|
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
