The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Nigeria » Children file sample size does not match with DHS Nigeria report (2008 and 2013)
Children file sample size does not match with DHS Nigeria report (2008 and 2013) [message #12950] Fri, 18 August 2017 15:16 Go to next message
trishasingh is currently offline  trishasingh
Messages: 5
Registered: August 2017
Member
There is a discrepancy between the sample size in DHS Nigeria reports and the current sample size of the associated children datasets.

For instance, the 2013 DHS report (pg. 177, Section 11.1.2.) mentions that 30,050 children were eligible for anthropometric measurements, while the raw DHS 2013 children file contains only 27,571 children between 0 and 59 months. (using the variable hw1 for determining age).

What could be a reason for this discrepancy?

[Updated on: Fri, 18 August 2017 15:16]

Report message to a moderator

Re: Children file sample size does not match with DHS Nigeria report (2008 and 2013) [message #13086 is a reply to message #12950] Sun, 17 September 2017 22:25 Go to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 626
Registered: January 2013
Senior Member
You are comparing the the wrong variable. hw1 is the age of children only for children of interviewed women. The anthropometry data are for all children in the household and are based on the de facto children listed in the PR dataset. If you try
tab hc1 if hv103==1
using the PR dataset you will find the 30,050 mentioned.
Previous Topic: Unmatched clusters when merging GPS data with DHS Nigeria datasets
Next Topic: Exclusive breastfeeding; clarification on the numbers.
Goto Forum:
  


Current Time: Tue Nov 13 00:41:40 Eastern Standard Time 2018