The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Using weights in regression analysis
Re: Using weights in regression analysis [message #73 is a reply to message #72] Wed, 20 February 2013 11:50
 Bridgette-DHS Messages: 3064Registered: February 2013 Senior Member
Here is a response from one of our DHS experts Tom Pullum, that should answer your question.

Future versions of the Guide to DHS Statistics will modify that recommendation.  Not using weights is a minority viewpoint here at DHS.  Almost all of us now advocate the use of weights.  How you use them will depend somewhat on your statistical package.  Most of us here use Stata.

If you do not use weights, the coefficients will be biased toward the over-sampled sub-populations.

For the HR and PR files, use hv005, for the IR, KR, and BR files, use v005, for the MR file use mv005.  The CR file contains both v005 and mv005.  It makes very little empirical difference which you use, but we prefer mv005 because it is adjusted for male non-response, which is typically more serious than female non-response.  Some people (e.g. Stan Becker) have proposed a composite couples weight, but as I said the effect of alternatives is trivial.

If you use the AR file, the weight is hiv05, and if you form a couples file using the AR data, the weight is hiv05 for males.

For some purposes it is convenient to divide the weights by 1,000,000, but in Stata, for example, pweight is unaffected by that, and for regressions you use pweight.

You also need to adjust for the clusters (the primary sampling units) and the strata.  In Stata that would be done with svyset and svy.  These adjustments do not alter the coefficients but they do alter the standard errors, usually in opposite directions.

I hope this helps.

Bridgette-DHS

[Updated on: Mon, 18 March 2013 09:11]

Report message to a moderator

 Using weights in regression analysis By: DHS user on Wed, 20 February 2013 11:48 Re: Using weights in regression analysis By: Bridgette-DHS on Wed, 20 February 2013 11:50 Re: Using weights in regression analysis By: enuanand on Tue, 19 March 2013 23:32 Re: Using weights in regression analysis By: Fabrice LOTY on Wed, 20 March 2013 05:34 Re: Using weights in regression analysis By: Traore on Wed, 20 March 2013 03:54 Re: Using weights in regression analysis By: Trevor-DHS on Wed, 20 March 2013 20:17 Re: Using weights in regression analysis By: idas on Fri, 29 March 2013 13:59 Re: Using weights in regression analysis By: Reduced-For(u)m on Sat, 30 March 2013 19:14 Re: Using weights in regression analysis By: idas on Tue, 02 April 2013 13:15 Re: Using weights in regression analysis By: Bridgette-DHS on Thu, 11 April 2013 17:31 Re: Using weights in regression analysis By: mnicolson on Fri, 14 June 2013 16:18 Re: Using weights in regression analysis By: Reduced-For(u)m on Fri, 14 June 2013 17:24 Re: Using weights in regression analysis By: smgwu on Thu, 17 October 2013 20:19 Re: Using weights in regression analysis By: Reduced-For(u)m on Sun, 20 October 2013 19:09 Re: Using weights in regression analysis By: myigzaw on Tue, 16 April 2013 08:25 Re: Using weights in regression analysis By: Reduced-For(u)m on Fri, 19 April 2013 01:04 Re: Using weights in regression analysis By: Bridgette-DHS on Fri, 26 April 2013 10:48 Re: Using weights in regression analysis By: Khaing Zar on Sat, 22 September 2018 02:31 Re: Using weights in regression analysis By: Bridgette-DHS on Mon, 24 September 2018 09:16 Re: Using weights in regression analysis By: kindu on Sat, 25 January 2020 12:09 Re: Using weights in regression analysis By: soumava on Wed, 07 February 2018 16:49 Re: Using weights in regression analysis By: Bridgette-DHS on Thu, 08 February 2018 09:40 Re: Using weights in regression analysis By: Khaing Zar on Thu, 20 September 2018 21:14
 Previous Topic: Deriving district population size from DHS weights? Next Topic: Interpretation of Rescaled household level weights for India-NFHS4
Goto Forum:

Current Time: Sat May 25 00:05:56 Coordinated Universal Time 2024