Project

General

Profile

Support #2099

Merging household info to individual data

Added by Marina Kousta 22 days ago. Updated 21 days ago.

Status:
Feedback
Priority:
High
Assignee:
Understanding Society User Support Team
Category:
Data documentation
Start date:
04/29/2024
% Done:

50%

Description

Hello,

I wish to perform some longitudinal analysis and I aim to append waves 9-13 from UKHLS. Before appending the waves, I wanted to first merge the corresponding wave's hhresp files to each indresp file of that wave. However, in most waves, it seems that after merging the hh info to the individual file, many observations stay unmatched both in my indresp and hhresp files. I attended an understanding society seminar a few years ago, where one of the exercises stated explicitly that although it is likely that from a certain HH no one filled any individual questionnaire in a given wave, it should not be the case that someone appears in the indresp file but there is no HH information on them from the hhresp file. I also found a document on the study design by Peter Lynn, which if I understood correctly, suggests that should be the case too. I am quite confused as with every wave I have tried, (using the merge command on stata as per the UKHLS suggested online syntax) I always get "unmatched" people for whom there is no HH info. I would appreciate any help in explaining this, whether it is potentially an error of the survey or an error of my code.

Many thanks in advance for your time.

Best wishes,
Marina

#1

Updated by Understanding Society User Support Team 21 days ago

  • Category changed from Data management to Data documentation
  • Status changed from New to Feedback
  • % Done changed from 0 to 50
  • Private changed from Yes to No

Hello Marina

Some households completed enumeration and at least one individual interview but did not respond to the household questionnaire. You can identify these cases by examining the “outcome” variable in hhsamp file, particularly looking for code 214. For Wave 9, when you merge indresp and hhresp files, you will find 731 observations that are present only in indresp but not in hhresp. All of these are categorised with code 214 in hhsamp file.

Here is an Stata code example for Wave 9 choosing variable “outcome” from hhsamp and “hsownd” from hhresp.
use "Y:\filepath\j_indresp.dta", clear
merge m:1 j_hidp using "Y:\filepath\j_hhsamp.dta", keepusing(j_outcome)
drop if merge==2
ren _merge merge
merge m:1 j_hidp using "Y:\filepath\j_hhresp.dta", keepusing(j_hsownd)
tab j_outcome _merge if _merge!=3

I hope this information is helpful.

Best wishes,
Roberto Cavazos
Understanding Society User Support Team

Also available in: Atom PDF