The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Nutrition and Anthropometry » DHS of India 2015/16 and 2005/06 (Calculation of stunting/underweight and overweight prevalence)
DHS of India 2015/16 and 2005/06 [message #17663] Fri, 03 May 2019 14:09 Go to previous message
StatMo is currently offline  StatMo
Messages: 4
Registered: May 2019
Dear all,

I am calculating the anthropometric failure prevalence for 25 countries with r-studio.
For 24 countries it worked fine. But for India I get illogical values.
I get an underweight prevalence for India 2016 of 20.8%(18.4%-23.1%) according to statcompiler it should be 35.7%.
When I calculate it manually then I get a value of 34.8%, however I need the standard error for my analysis
so I have to consider the complex study design with the package survey in R. I get the error message: Error in onestrat(x[index, , drop = FALSE], clusters[index], nPSU[index][1], :
Stratum (5610) has only one PSU at stage 1 , then I solve it with options(survey.lonely.psu= "remove") I also tried adjust and average, but it always gives me the same result,
only with a different standard error. I don't know what to do :-/

There might be something data specific I did not consider because the same code worked perfectly fine for the other countries.
I hope someone of you understands my problem!


[Updated on: Sat, 04 May 2019 10:36]

Report message to a moderator

Read Message
Read Message
Previous Topic: M55 variable missing
Next Topic: incorrect stunting rates
Goto Forum:

Current Time: Sun Feb 23 12:49:34 Eastern Standard Time 2020