The DHS Program User Forum: Other countries » Household Respondent's age (Philippines, years 2013,2017 and 2022)

Home » Countries » Other countries » Household Respondent's age (Philippines, years 2013,2017 and 2022)

Show: Today's Messages :: Show Polls :: Message Navigator

Household Respondent's age (Philippines, years 2013,2017 and 2022) [message #27121]

Sun, 18 June 2023 06:54

josueestrada@ph
Messages: 1
Registered: June 2023

Member

Hello,

I am doing a research involving data from the Philippine DHS for the years 2013, 2017, and 2022. As my analytic sample, I will be including the respondents to the houshold survey. I am trying to determine their age but when I prepare the dataset for my analytic sample, I get respondents with ages below 15 and as high as 96+. I also have a few respondents with unknown ages. I've tried creating the dataset both from the household (HR) and individual members (PR) datasets, matching/keeping information based on hv003 (line number of each respondent) for each household number on a cluster and for a particular year. As an example, here are the commands I used in Stata to extract the information in the PR datasets:

bysort hv007 hv001 hv002: gen seq=_n
bysort seq: keep hvidx==hv003

Can someone help me figure out if I did something wrong? I would appreciate your help. Thank you very much.

Kind regards,
Josue

Report message to a moderator

Re: Household Respondent's age (Philippines, years 2013,2017 and 2022) [message #27216 is a reply to message #27121]

Fri, 30 June 2023 16:03

Janet-DHS
Messages: 698
Registered: April 2022

Senior Member

Following is a response from DHS staff member, Tom Pullum:

We apologize for the delay in this reply. You seem to be correct that there is a much wider range of ages for household respondents than there should be. I used a different strategy,. I opened the 2017 PR file, and then entered the following lines:

keep if hvidx==hv003
keep hv001 hv002 hv104 hv105

rename hv104 respondent_sex
rename hv105 respondent_age

tab respondent_age respondent_sex
sort hv001 hv002

This gives a file that can be merged back with the PR file. I get a few household respondents below age 15, which I would not have expected. There are other ways to do this, with the HR file (as you say) as well as with the PR file.

Report message to a moderator

Previous Topic:	Cambodia DHS (all years) - Total chacracteristic not available from the API endpoint
Next Topic:	Cambodia DHS (all years) - Regional labels appear in duplicate and triplicate

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Apr 27 02:59:04 Coordinated Universal Time 2024