Support #2093
open
Added by Felicity Leung 5 months ago.
Updated 5 months ago.
Category:
Questionnaire content
Description
Hello,
For Wave 10,11 and 12, why does the variable "voteeuref" have such a high proportion of inapplicable and missing observations?
I think in the Universe section of the questionnaire it says the EU referendum module was only asked on certain dates? Which dates were these and why?
Thank you in advance.
- Category set to Questionnaire content
- Status changed from New to Feedback
- % Done changed from 0 to 80
- Private changed from Yes to No
Hi Felicity,
The reasons for the missing data you're observing are quite complex.
First, you need to check the w_euref variable as it filtered the w_voteeuref one. Second, due to an error these questions weren't asked during interviews in sample year 1 (2018) in wave 10 (you can check that for j_istrtdaty=2018 j_euref=-9). Third, in waves 11 and 12 these questions were also not asked in all months (this time intentionally). In wave 11 political behaviour content was added midway in Wave 11 to capture the European Elections in May 2019 and the General election in December 2019, so we started asking the questions around early June 2019 and continued until the end of the fieldwork (May 2021). In wave 12 these questions were asked from the beginning of the fieldwork in January 2020 up to late February 2021.
Also, note that due to the combination of an unintended error and a complex fieldwork setup, there are some respondents who answered these questions more than once, so you need to check the consistency of the answers for such cases.
I hope it helps.
Best wishes,
Piotr Marzec
UKHLS User Support
Also available in: Atom
PDF
Updated by