Support #1285

Missing interview year and month

Added by C Josten 8 months ago. Updated 8 months ago.

Data documentation
Target version:
Start date:
Due date:
% Done:


Estimated time:


I realised that the variable interview month and year is missing for all of BHPS wave one and for parts of other waves (e.g. UKHLS wave 1). Is there a reason for this? And can one assume the year of the interview to be the year of the wave or should one ignore those observations from the analysis?

Thanks and kind regards,


#1 Updated by Stephanie Auty 8 months ago

  • Private changed from Yes to No
  • % Done changed from 0 to 50
  • Assignee set to Stephanie Auty
  • Status changed from New to In Progress
  • Category set to Data documentation

Dear C,

BHPS does not include an interview year variable because all of the interviews took place within one calendar year, so yes, please use the year of that wave of the study. I can see the interview month variable in ba_indresp and only 21 are missing.

There are some missing values for interview dates in UKHLS, and in Wave 1 these all relate to proxy interviews. I will look in to why this may be the case.

Best wishes,

#2 Updated by Stephanie Auty 8 months ago

  • % Done changed from 50 to 80
  • Assignee changed from Stephanie Auty to C Josten
  • Status changed from In Progress to Feedback

Dear C,

For Understanding Society datasets, please use the derived date variables, w_intdatd_dv w_intdatm_dv w_intdaty_dv, as these have been imputed where possible. The note for the year variable is as follows:

Date of interview, year. For adult respondents this is the date their personal interview took place. If all interviews in the household took place on the same day, all non-respondents (and youth respondents for whom no interview date is recorded) received the other household members' interview date. For the remaining cases, children under the age of 12 received the interview date of the oldest female in the household (who in most cases will be the mother and provide information about the child as part of her interview). Following this, a random interview date in the household is chosen. To compute derived interview dates, crosswave inconsistencies in recorded interview dates were resolved.

Best wishes,

Also available in: Atom PDF