Support #2307
open
Questions about 'pensioner_dv' and 'pensioner'
50%
Description
Dear UKHLS Team,
I am using Understanding Society (UKHLS) data to select samples of individuals who are of State Pension age at the time of the survey.
First, I have noticed that the variables pensioner_dv and pensioner produce different results, even within the same wave. Could you please clarify the differences between these two variables?
Second, the description for pensioner states: 'Respondent is of pensionable age: Whether the respondent is of state pensionable age or not at the time of the survey is calculated using the government defined rules in force based on their date of birth or age.' Does this mean it uses the rules that were in force at the * time of the survey* ?
Third, the description for pensioner_dv states: 'Whether R is of pensionable age: Whether respondent is of state pensionable age. Uses derived age and interview information and state pension rules applicable since 6 April 2016'. Does this mean that the post-2016 rules are applied retrospectively to data from earlier waves?
Fourth, for the purpose of accurately identifying whether a respondent was eligible for the state pension (i.e., at or above State Pension age) at the time of each survey interview , would it be appropriate to use one of these variables consistently across waves? If so, which one would you recommend?
Thank you very much for your assistance—I greatly appreciate your help in clarifying these points.
Best regards,
Bingqing
Updated by Understanding Society User Support Team 4 days ago
- Status changed from New to Feedback
- % Done changed from 0 to 50
- Private changed from Yes to No
Hello Bingqing,
The variables pensioner_dv and pensioner produce different information mainly because *_dv variables are created after the survey fieldwork has taken place. As specified in the note you mention, pensioner_dv uses age_dv, which is also a post-fieldwork derived variable. In contrast, pensioner is created by the fieldwork agency at the time of the interview and is part of the grid questionnaire module.
Regarding your second and third questions, the answer is yes.
As for your fourth question, it depends on what exactly you want to measure. One advantage of using the derived variable is that it has already gone through a cleaning and harmonisation process, so the resulting classification is more accurate than pensioner. However, the rules applied correspond to those in place from 6 April 2016. Therefore, if you want to capture the exact pensioner classification according to the rules in force at the time of each interview, you would need to use pensioner or construct the measure yourself.
I hope this information is helpful
Best wishes,
Roberto Cavazos
Understanding Society User Support Team