Weights for use in sub-population [message #11262] |
Sat, 26 November 2016 06:33 |
Mekdes
Messages: 1 Registered: November 2016 Location: Gent
|
Member |
|
|
I am working on a data that has a sub-sample of women in reproductive age group (20-49 years, non-pregnant, and non-lactating). I am not sure if it is correct for me to use the weights provided by the DHS or compute a new weight for my target group?
To calculate a new weight, I think I need more data that are not immediately available on DHS program website - such as;
- total number of households per each cluster,
- selected number of households per each cluster,
- number of households interviewed in each cluster,
- total number of women selected for interview in each cluster and
- number of eligible women interviewed in each cluster.
If it is required for me to calculate a new weight, where can I look for the required data (listed above)?
|
|
|
Re: Weights for use in sub-population [message #11278 is a reply to message #11262] |
Mon, 28 November 2016 08:15 |
Bridgette-DHS
Messages: 3214 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
Quote:When you use the weights (v005, etc.), the sample is representative of the full population. Sub-sampling the sample is then equivalent to sampling from the full population. The sub-sample you describe, with weights, will be representative of the sub-population you want to describe. There is no need for re-weighting.
|
|
|