The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Egypt » issue with Husband/partner age (IR recode file )
issue with Husband/partner age [message #27725] Tue, 26 September 2023 21:57 Go to next message
Alanood is currently offline  Alanood
Messages: 18
Registered: November 2022
Member
Hi


I am using EGIR5ADT EGIR61DT, which are 2008 - 2014 IR recode files for Egypt.

I constructed the age gap in STATA by this command

gen age_gap= v730. - v012

where
v730 husband/partner age
v012 mother age

however, some results that I got are not matched with other variable.
for example, which pasted bellow, A man aged 15 will not be able to have higher eduaction in his age or having 7 children.

Husage: 15
mage: 49
age_gap: -34
nofchildern: 7
wealth: Richest
m_occ: Not employed
hus_occ: Professional, technical or managerial
hedu_years: 15
heducation: Higher education



where
Husage is the husband age
mage is the mother age
nofchildern is number of children under 5 in the HH
m_occ id mother occupation
hus_occ is husband occupation
hedu_years is husband education in years
heducation is husband educational level





This issue found in couple of individuals in the sample and my question here, Is this count as outliers in the dataset and need to be dropped ?




Thank you..



[Updated on: Wed, 27 September 2023 01:33]

Report message to a moderator

Re: issue with Husband/partner age [message #27728 is a reply to message #27725] Wed, 27 September 2023 09:15 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3172
Registered: February 2013
Senior Member

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

There are probably data quality issues with the woman's report of her husband's age. In the IR61 file I see that the difference ranges from -34 to +68. It's hard to say what combinations are implausible, because there probably are genuine cases of large age differences, although more likely in a positive rather than negative direction. Your checks for education and children ever born are helpful. You could also look at the MR and CR files EXCEPT THAT Egypt has not included interviews with men, so those files do not exist.

Note that the woman may have had an earlier marriage and there may be other wives (see v505).

I suggest that you categorize the age difference so you have open-ended categories for large negative and large positive differences, and in your analysis you comment on the implausibility of the extreme values. I would NOT recommend that you try to adjust or edit the individual values of the husband's age.
Previous Topic: Searching Variable of Governorates
Next Topic: 2015 dataset, female genital cutting
Goto Forum:
  


Current Time: Thu Oct 10 19:17:24 Coordinated Universal Time 2024