| Re: District-level aggregates [message #12333 is a reply to message #12332] |
Tue, 02 May 2017 15:05   |
Bridgette-DHS
Messages: 3199
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
DHS surveys are representative at the stratum level, which is all the combinations of regions and urban/rural places. In order to get reliable estimates at the stratum level, there is a tendency to over-sample the smaller strata (and then weight them down) and under-sample the larger strata (and then weight them up). You are right, that for districts the problem is with confidence and power, rather than with bias. The expected values of estimates for smaller errors should be the population values, but confidence intervals of the estimates will be wider and the probability of identifying a real difference is smaller. It's always a good idea to look at the standard errors of the estimates, no matter what level you work at.
Displacement of clusters is always within districts, as well as within regions, so that does not affect district-level estimates.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
