The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » Age in months: Missing values
Age in months: Missing values [message #6689] Sat, 27 June 2015 20:03 Go to next message
MissJibo is currently offline  MissJibo
Messages: 5
Registered: May 2015
Location: United Kingdom
Member
I am working on DPT3 immunization coverage among 8 different countries and wish to use children aged 4 months to 23 months as denominators. Unfortunately, in most countries, more than 50% of the variable "age in months" is missing. Is there a systematic reason why this is so? Anyone else has this problem?

P.S. I asked this question in the general forum and was told to look at the DHS guide to statistics which I did. But there is no explanation for this in there. Thank you.

Re: Age in months: Missing values [message #6690 is a reply to message #6689] Sat, 27 June 2015 22:08 Go to previous messageGo to next message
Reduced-For(u)m
Messages: 292
Registered: March 2013
Senior Member

There are some people who don't know the exact birthdate of their children, but 50% seems too high - which variable and recode are you using? If you look at "age in years" for those missing "age in months", are they actually very young children?
Re: Age in months: Missing values [message #6691 is a reply to message #6690] Sat, 27 June 2015 22:32 Go to previous messageGo to next message
MissJibo is currently offline  MissJibo
Messages: 5
Registered: May 2015
Location: United Kingdom
Member
Hi. Thanks for the reply. I am using DHS data from Namibia, Nigeria, DRC, Philippines, Liberia, Sierra Leone, Mali and Dominican Republic. Most have up to 20%-50% missing data for that variable.I just checked to see those with missing variables, and they are equally distributed among all ages in years. Thank you very much.
Re: Age in months: Missing values [message #6794 is a reply to message #6691] Fri, 17 July 2015 10:50 Go to previous messageGo to next message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
Here is a response from one of our technical experts, Dr. Tom Pullum:
Quote:
You are apparently talking about age at immunization, rather than age. Age in months is given as hw1 for all surviving children age 0-4 years. The KR file includes children born in the past 5 years who died, as well as those who survived. If the child died (b5=0) then hw1 is missing and there is also no immunization data.

Age at immunization can only be calculated if we have the date of immunization, which requires a health card with a date. That's given as h7=1. If h7 has any other code, then h7d, h7m, and h7y will be missing and you cannot calculate date of immunization or age at immunization.

There are other children who are taken to be immunized because the mother said so, but for them, no date is recorded. If there were a date, it would not be reliable.

I will paste below some Stata lines for calculating age at immunization, in months, for those children with non-missing date.

set more off
tab1 h7*

gen h7d_rev=h7d
replace h7d_rev=. if h7d>31

gen h7m_rev=h7m
replace h7m_rev=. if h7m>12

gen h7y_rev=h7y
replace h7y_rev=. if h7y>v007

gen h7_mdy=mdy(h7m_rev,h7d_rev,h7y_rev)

gen dob_mdy=mdy(b1,hw16,b2)

* "exact" age in months; the average days in a month is 365.25.12 = 30.4375
gen h7_age_in_months=(h7_mdy-dob_mdy)/30.4375
summarize h7_age_in_months 

* completed age in months
gen h7_completed_age_in_months =int(h7_age_in_months)

summarize h7_age_in_months
tab h7_completed_age_in_months


[Updated on: Fri, 17 July 2015 14:46]

Report message to a moderator

Re: Age in months: Missing values [message #8700 is a reply to message #6794] Thu, 03 December 2015 06:58 Go to previous message
Salam is currently offline  Salam
Messages: 1
Registered: November 2015
Location: Johannesburg
Member
Thanks for this code.

Dr Salam Mbengue
Previous Topic: Dominican Republic 2007 child immunization
Next Topic: Using svy: tab
Goto Forum:
  


Current Time: Thu Dec 5 12:20:10 Coordinated Universal Time 2024