Project

General

Profile

Actions

Support #2164

open

Merging w_hhresp and w_indresp vars as a function of PNO

Added by Emma Bridger about 2 months ago. Updated 3 days ago.

Status:
Feedback
Priority:
Normal
Category:
Data management
Start date:
10/19/2024
% Done:

80%


Description

Hello,

Apologies if this is already covered but I haven't been able to find it in the FAQs and user guide. I am looking to merge variables from w_hhresp and w_indresp initially within-wave but eventually across waves. For my research question it is important that I merge only data from w_indresp that corresponds to the exact person who completed the w_hhresp telephone interview, or at least that I code which individual pidp completed the hh survey. I think this information is contained in the w_ivh1 - w_ivh16 variables but would like to (1) confirm whether this is correct and (2) ask whether there is any guidance or syntax for doing this correctly?

More generally, I would also like to confirm that I am correct in my understanding that the household grid question were completed only once by one member of the household on behalf of all other members.

Thanks in advance,
Emma

Actions #1

Updated by Understanding Society User Support Team about 1 month ago

  • Status changed from New to In Progress
  • Private changed from Yes to No

Dear Emma,

Thank you for your question. We’ve contacted our fieldwork expert about the exact procedure and will get back to you once we have more information.

Best wishes,
Piotr Marzec
UKHLS User Support

Actions #2

Updated by Understanding Society User Support Team about 1 month ago

  • Category set to Data management
  • % Done changed from 0 to 80

Dear Emma,

You are correct about the w_ivh1 - w_ivh16 variables. While we don’t have specific syntax for this situation, you can refer to the following resources on our website https://www.understandingsociety.ac.uk/documentation/mainstage/syntax/

Distributing household-level information to the individual level
Summarising individual-level information at the household level
Merging individual files across waves into wide format
Merging individual files across waves into long format

Your linking variable will be the combination of w_hidp and the PNO from w_ivh1 - w_ivh16.

It is possible for more than one person to answer the household questionnaire. This usually occurs in face-to-face interviews, where one member of a couple may know more about certain topics (e.g., mortgage/rent), while the other knows more about expenses like food and utility bills. That’s why there is a set of binary variables.

I hope it helps.

Best wishes,
Piotr Marzec
UKHLS User Support

Actions #3

Updated by Understanding Society User Support Team about 1 month ago

  • Status changed from In Progress to Feedback
Actions #4

Updated by Emma Bridger 26 days ago

Understanding Society User Support Team wrote in #note-2:

Dear Emma,

You are correct about the w_ivh1 - w_ivh16 variables. While we don’t have specific syntax for this situation, you can refer to the following resources on our website https://www.understandingsociety.ac.uk/documentation/mainstage/syntax/

Distributing household-level information to the individual level
Summarising individual-level information at the household level
Merging individual files across waves into wide format
Merging individual files across waves into long format

Your linking variable will be the combination of w_hidp and the PNO from w_ivh1 - w_ivh16.

It is possible for more than one person to answer the household questionnaire. This usually occurs in face-to-face interviews, where one member of a couple may know more about certain topics (e.g., mortgage/rent), while the other knows more about expenses like food and utility bills. That’s why there is a set of binary variables.

I hope it helps.

Best wishes,
Piotr Marzec
UKHLS User Support

Hi Piotr,

That definitely helps - thank you very much. When I'm looking at these w_ivh1 - w_ivh16 for the specific wave I'm looking at, the questionnaire appears to have been completed by a non-household member in nearly 5% of cases. Is there a way to determine who this person might be? Seems like a larger proportion than I would expect.

Two other quick questions:
1. Can I just check that the PNO referenced in w_ivh1 - w_ivh16 (in w_hhresp) maps cleanly onto the pno variable in w_indresp.
2. I am interested in the Matdep hh_resp questions. I can see that these were asked only when "If ((GRIDVARIABLES.NKIDS015 > 0) | (GRIDVARIABLES.NKIDS015 = 0 & GRIDVARIABLES.NPENSIONER = 0))". Do I understand this correctly as only asked when "One or more children aged 0 to 15 in the household OR no children aged 0 to 15 in the household AND nobody of pensionable age in the household"? Why was this not asked of all households? Am I correct in interpreting therefore that these were only shown to household with nobody of pensionable age, and the separate pdep questions were presented to the remaining households? Is there any documentation to this?

Many thanks,
Emma

Actions #5

Updated by Understanding Society User Support Team 24 days ago

Hi Emma,

1) Yes.
2) Yes, you're correct about the universe for the matdep questions – these are not asked of households with pensioners, and there is a separate set of pdep questions for those households. The former is adapted from the Family Resources Survey, while the latter originates from the Department for Work and Pensions. Looking at the questionnaires, it seems that the matdep questions were asked of everyone in waves 1 and 2, but a more limited routing was introduced from wave 4 onwards. I’ll try to find out why this change was made and will get back to you when I have more information.

Best wishes,
Piotr Marzec
UKHLS User support

Actions #6

Updated by Emma Bridger 3 days ago

Hello,

Is there any documentation about the decision making about which questions to include in the household vs. individual-level questionnaire, particularly for wave 12?
I am looking at wave 12 and can't see anything in the technical fieldword report about this. I am particularly interested in the development of the questions on "Neighbourhood Conditions" (waves 3, 6, 9, 12) and why these were allocated to the household questionnaire whereas the items on "local neighbourhood" service quality and trust were incorporated in the individual level questionnaire.

Many thanks,
Emma

Actions

Also available in: Atom PDF