| Re: Mismatched Child Counts in Stunting Analysis vs. DHS Rwanda Report [message #30335 is a reply to message #30272] |
Thu, 07 November 2024 21:03  |
Janet-DHS
Messages: 888
Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS staff member, Tom Pullum:
We apologize for the delay in this response. Table 11.1 is not calculated from the KR file. It is calculated from the PR file, using children under 5 (cases in the PR file with hc1<60) and hc70, hc71, and hc72 for the HAZ, WAZ, and WHZ, respectively and weighted by hw005. The table is limited to de facto residents (children with hv103=1, who slept in the household the night before).
The following lines reproduce n=4158 (rounded from 4157.661) in the final report:
gen wt=hv005/1000000
total wt if hc70<9994 & hv103==1
Total estimation Number of obs = 4,049
------------------------------------------------------------ --
| Total Std. Err. [95% Conf. Interval]
-------------+---------------------------------------------- --
wt | 4157.661 24.87947 4108.884 4206.439
------------------------------------------------------------ --
If you want to relate stunting, etc, to covariates that are in the KR file but not the PR file, then you would use the KR file.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
