The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » weighting data
weighting data [message #757] Wed, 11 September 2013 14:38 Go to next message
jn2255 is currently offline  jn2255
Messages: 2
Registered: September 2013
Member
Hi,

What is the appropriate stata command for weighting the Bangladesh 2007 data? I want to weight the child recode file but cant figure out what is the appropriate strata variable for that dataset.

Also, if I use the svyset command at the beginning of using the dataset, must I repeatedly specify "svy" in every subsequent stata comman in which I wish for stata to treat it like a population instead of subsample?


Thanks.
Re: weighting data [message #1414 is a reply to message #757] Sun, 23 February 2014 17:45 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
1) You can use the following to set up your survey design, stratification and weighting:

gen psu=v021
gen strata=v023
gen weight=v005/1000000
svyset psu [pweight=weight], strata(strata)

Note that the strata can vary from survey to survey and may not always be v023. In many surveys the strata are urban and rural areas within each region and can be produced with
egen strata = group(v024 v025), label
However, for the Bangladesh 2007 this would be incorrect as the strata are not just urban and rural areas in each region. See Appendix A of most final reports for information on the stratum used in selecting the sample. In this survey, v023 contains the 22 strata used for the stratification of the sample.

2) Yes, you should use svy in the subsequent commands.
Re: weighting data [message #3512 is a reply to message #1414] Sun, 21 December 2014 05:46 Go to previous messageGo to next message
lhuri is currently offline  lhuri
Messages: 16
Registered: December 2014
Location: Indonesia
Member
Dear all,

I'd like to know which variables in special DHS (I'm working on Indonesia Adolescent Reproductive Health) to use for weighting purpose.
Because I notice that they are different from those of standard DHS. I find "v0005" equal to "aweight" in special DHS. But I'm not sure the equal variables for PSU and strata. Does anyone here ever use special DHS dataset and know how to properly use sample weights?
Thank you,
Lhuri
Re: weighting data [message #3527 is a reply to message #3512] Fri, 26 December 2014 11:21 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
gen psu=acluster
egen strata=group(aprov atype)
gen weight=aweight/1000000
svyset psu [pweight=weight], strata(strata)
Re: weighting data [message #9822 is a reply to message #757] Thu, 26 May 2016 03:29 Go to previous messageGo to next message
Alfi Lailiyah is currently offline  Alfi Lailiyah
Messages: 2
Registered: May 2016
Location: Indonesia
Member
Dear All

I am using BR and MR data, to know the factors related to pneumonia. Which weight must I use? BR's weight or MR's weight? The variable in MR that I need is smoking status of father

Thank you,
Re: weighting data [message #9911 is a reply to message #9822] Mon, 06 June 2016 08:09 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member
Following is a response from Senior DHS Data Processing Specialist, Noureddine Abderrahim:


The use of sample weights is described on the DHS program website. Follow the link below and check "Step 7":

https://www.dhsprogram.com/data/Using-DataSets-for-Analysis. cfm#CP_JUMP_14042

Previous Topic: Weighting in pooled data
Next Topic: Weighting Data for Pooled Indonesian DHS Dataset
Goto Forum:
  


Current Time: Thu Mar 28 08:56:02 Coordinated Universal Time 2024