The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » Why are there fewer men among 15-49 year olds? (Does the adverse proportion suggest selection bias?)
Why are there fewer men among 15-49 year olds? [message #15926] Sat, 06 October 2018 18:11 Go to next message
nibiti is currently offline  nibiti
Messages: 20
Registered: April 2018
Location: Göttingen
Dear all,

I have compiled a DHS dataset of Sub-Saharan African countries to analyze several anthropometric health indicators (birth weight, WAZ, HAZ, adult height). The dataset is a combination of PR, IR, MR and KR. I want to analyze effects of conflict on both men and women.

Now, I noticed that the number of women (age 15-49) in the sample is much larger than for men in this age group.

My concern is, that this reflects a selection bias. For example, it could be possible that there are fewer men in this sample, because men were often not home during the DHS interviews. This would mean that men that were home during the interview could be systematically different from those who were not present during interviews (e.g. because they were not suited for agricultural work).

However, I think that the adverse proportion of women to men could have also been an active decision of USAID.

Maybe you could tell me, if that is the case and where I could find more information on this issue (so that I can cite the information I get)? Are the men in the sample representative for all men in the countries?

Many thanks in advance,
Re: Why are there fewer men among 15-49 year olds? [message #16029 is a reply to message #15926] Wed, 24 October 2018 10:36 Go to previous message
Messages: 1516
Registered: February 2013
Senior Member
Dear User, a response from senior sampler, Dr. Ruilin Ren:

Male surveys are usually conducted in a sub-sample of households. the appendix A should provide this information.
I think the best place to start is to look at the sampling section of each final report. There is an Appendix, generally Appendix A which deals with the sample design. Every survey is unique so it is worth looking at this section. After reviewing, if you have additional questions, please feel free to post again. Thank you!

[Updated on: Wed, 24 October 2018 22:03]

Report message to a moderator

Previous Topic: individual and community factors of PNC utilization among mother's using 2016 Edhs
Next Topic: Unable to provide estimates of population health indicators using Geo codes in India
Goto Forum:

Current Time: Fri Feb 23 22:03:59 Coordinated Universal Time 2024