The DHS Program User Forum: Weighting data » Weighting in regional subsample

Home » Data » Weighting data » Weighting in regional subsample

Show: Today's Messages :: Show Polls :: Message Navigator

Weighting in regional subsample [message #8151]

Sat, 29 August 2015 06:17

christian-spi
Messages: 1
Registered: August 2015

Member

I am working with the latest DRC questionnaire. I am performing both a descriptive and a regression analysis on a dataset where I have merged the INDIVIDUAL HOUSEHOLD with the WOMEN file.

1. Which weights should be used on the national sample? Would I use women weights whenever I am interested in variable from the female questionnaire and individual household weights if the variable of interest is on the household level?

2. I would like to perform an analysis for a regional subsample which comprises women between 15-30 in Kinshasa. Which weights would I use for the subsample analysis?

Thank you very much for your help!

Best wishes

Report message to a moderator

Re: Weighting in regional subsample [message #8170 is a reply to message #8151]

Tue, 01 September 2015 12:09

Bridgette-DHS
Messages: 3214
Registered: February 2013

Senior Member

Following is a response from Senior Data Processing Specialist, Noureddine Abderrahim:

- The file you constructed is a file of individuals, to produce statistics of variables in this file, you will have to use the individual weight.

- For the subsample, you need to use the individual weight as well. However, you need to make sure that the number of cases is not too small.

Report message to a moderator

Previous Topic:	NDHS 2011
Next Topic:	Child Weights

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Dec 25 02:16:09 Coordinated Universal Time 2024