Project

General

Profile

Support #631

Rents in Understanding Society

Added by Maria Sanchez-Vidal over 7 years ago. Updated over 7 years ago.

Status:
Closed
Priority:
Normal
Category:
Data inconsistency
Start date:
09/22/2016
% Done:

100%


Description

I am using the Secure version of Understanding Society and I have realised that the rent variable (w_rent) is somehow inconsistent. As an example, I have homogenised the variable in order to have the weekly rent but, when comparing the same person across waves, the variable could easly be £10,000/week in wave a, £100/week in wave b and £880/week in wave c, even if the coordinates have not changed across waves (i.e. live in the same place).

What is the reason for this to happen?

Thank you in advance,

Maria

#1

Updated by Victoria Nolan over 7 years ago

  • Status changed from New to Feedback
  • Assignee changed from Alita Nandi to Maria Sanchez-Vidal
  • % Done changed from 0 to 50
  • Private changed from Yes to No

Dear Maria,

We just want to confirm that:
(i) you are using the variable, w_origadd, to identify whether the respondent/hh is still living at the same address as last wave: https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/wave/2/datafile/b_hhresp/variable/b_origadd
(ii) you are dividing the variable b_rent by b_rentwc to compute the weekly rent: https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/wave/2/datafile/b_hhresp/variable/b_rent
https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/wave/2/datafile/b_hhresp/variable/b_rentwc

If you are using these variables, and still find these unusual rent changes then could you please let us know the frequency of such large variations. If there is a very high proportion of such cases, we will investigate to see if there are any systematic data errors. If not, these could simply be some random data entry/data reporting errors. For example, the case that you have discussed could easily have been data entry errors where extra 0s were added. So, these could have been £100/week in wave a, £100/week in wave b and £88 in wave c. In such cases, researchers often use other information to obtain an estimate. For example, you could check the council tax band to see which one of these is possibly an error (b_rentinc3). Alternatively, different persons may have answered these questions in each wave, and one or more of them may have misunderstood the question. You can see who answered these questions from the data (w_ivfh1-w_ivfh16).

Please note, we refer to a generic wave prefix using w.

Best wishes, Victoria

On behalf of the Understanding Society Data User Support Team

#2

Updated by Victoria Nolan over 7 years ago

  • Status changed from Feedback to Closed
  • % Done changed from 50 to 100

Also available in: Atom PDF