The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » HIV » Cameroon: Sample domain missing for weights
Cameroon: Sample domain missing for weights Thu, 12 December 2013 13:42
 ab1506 Messages: 1Registered: December 2013 Location: London Member
Hi all

I hope you can help me with this problem. I am trying to get HIV data for couples from the Cameroon DHS but the sample domain variable used for sample weights is empty (mv023). I tried using mv101 (region) instead as I noticed that for Malawi mv023 and mv101 were the same but I don't get exactly the same results for HIV prevalence in couples as in their DHS report so I am guessing mv101 might not be the way to go.

I get out of 2875 observations:
----------+------------
missing | .0132
concordant negative | .9139
discordant F positive| .031
discordant M positive| .0275
concordant positive | .0144
Total | 1

The DHS report gives out of 2871 observations :
concordant negative 1,5
discordant F positive 3,1
discordant M positive 2,8
concordant negative 92,6

I can see the denominator is not the same but I am using the same merging code for couples + HIV that I have used with other countries successfully so I think the problem might come from the sample domain used.

Annick

I am an HIV epidemiologist interested in marginalized populations and social determinants of health and mostly do mathematical modelling

[Updated on: Fri, 13 December 2013 04:10]

Report message to a moderator

Re: Cameroon: Sample domain missing for weights [message #1127 is a reply to message #977] Fri, 10 January 2014 06:44
 Liz-DHS Messages: 1516Registered: February 2013 Senior Member
Dear Annick,
There are two variables used which may be of interest to you. v101w Region and stype1w Milieu de résidence (described below)
I am not a programmer but I looked at some of the code used to create Table 15.12 P284-285 in the final report(in CSPro)titled "Prevalence among couples" and found the following:
crosstab float(1) t1412 v013w7+v013w8+agedif+v505w2+multpart+concpart+stype1w+v101w+ v106w1+v106w2+v190w+total col1412+totnum1
exclude(rowzero,colzero,percents,totals,specval)

title( "Table 14.12 HIV prevalence among couples","",
"Percent distribution of couples living in the same household, both ",
"of whom were tested for HIV, by HIV status, according to ",
"background characteristics, Country 2011" )
stub( "Background characteristic" );

The row variables are:
v013w7 Woman's age
v0138w8 Man's age
agedif Age difference between partners
v505w2 Type of union
multpart Multiple partners in past 12 months
concpart Concurrent sexual partners
stype1w Miliew de résidence with 1 Yaoundé/Doula 2 Autres villes 3 Ensemble urbain 4 Rural
v101w Region
v106w1 Woman's education
v106w2 Man's education
v190w Wealth quintile
total Total

I am including the link to the final report. I believe you were referring to Chapter 15, Table 15.12 on pages 84-85.
http://www.measuredhs.com/pubs/pdf/FR260/FR260.pdf