The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » India » Discrepancy in Estimates using NFHS5
Discrepancy in Estimates using NFHS5 [message #24624] Thu, 09 June 2022 07:10
aarushikataria is currently offline  aarushikataria
Messages: 1
Registered: June 2022
Member
Hello,

I was trying to obtain an estimate of the proportion of women that consume alcohol using the NFHS5 Individual Recode. The below is the code I was running on STATA:
sampleweight = v005/1000000
rename (s720 s721) (wifealc wdrinkfreq)
egen strata = group(v024 v025), label
* tab strata
svyset [pweight=sampleweight], psu(v021) strata(strata)
svy: tab wifealc

This gives me that 0.74% of women consume alcohol. However, the DHS and NFHS5 report say this number is 1.3%. On closer inspection, I found that the DHS report (https://www.dhsprogram.com/pubs/pdf/FR375/FR375.pdf) notes that the total number of observations in this dataset are 1,044,919 to arrive at 1.3%. However, the Individual recode has only 724,115 observations. Could the difference in observations be the reason for discrepancy?

Would be very grateful for any leads.
Previous Topic: Individual Recode Data
Next Topic: Caste waves 2015-2019
Goto Forum:
  


Current Time: Fri Mar 29 05:33:08 Coordinated Universal Time 2024