Support #1131

Harmonization of retrospective cohabitation and marital status information

Added by Ashley Burdett over 5 years ago. Updated over 3 years ago.

Start date:
% Done:



Dear UKHLS User Support,

I am currently trying to compile complete relationship histories for individuals using the retrospective cohabitation and marital status files from both the BHPS and USoc. However, I am having difficulty assessing how compatible the retrospective data files are.

To offer an example, some of the variables appear very similar, e.g. cohabitation start dates in the retrospective datasets in wave b BHPS and wave 1 USoc, but the BHPS variables have the bw_ prefix and the _bh suffix. It is unclear to me why this is the case.

Further, it appears that the retrospective marriage dataset in USoc groups marriage and civil partnership together, whereas, given civil partnership was not possible until 2004, the equivalent dataset in wave b of the BHPS only contains information for marriage. It would be useful to know if this is the only reason that the two datasets are not equivalent, thus if I am willing to group together marriage and civil partnership I would be able to make them equivalent or not.

Many thanks in advance.



Updated by Stephanie Auty over 5 years ago

  • Status changed from New to In Progress
  • % Done changed from 0 to 10
  • Private changed from Yes to No

Many thanks for your enquiry. The Understanding Society team is looking into it and we will get back to you as soon as we can.

Best wishes,
Stephanie Auty - Understanding Society User Support Officer


Updated by Stephanie Auty over 5 years ago

  • % Done changed from 10 to 50

Dear Ashley,

All of the variables and data files from BHPS have a bw_ prefix in the harmonised data, that is how we distinguish Wave 1 of BHPS from Wave 1 of UKHLS.

During harmonnisation a variable was given a _bh suffix if it had the same name as a UKHLS variable but was different in some way. This could be in the question wording or response options. You can check this by reading the questionnaires:

I am waiting for a response regarding the last part of your question, apologies for the delay. Reading the questionnaires may also help with this in the meantime.

Best wishes,


Updated by Alita Nandi over 5 years ago

  • Assignee changed from Stephanie Auty to Ashley Burdett
  • % Done changed from 50 to 70

Yes, you are correct about civil partnership not being asked in the BHPS retrospective marriage histories but grouped together in UKHLS. Also note that the current marital status at each interview (w_mastat_dv) distinguishes between marriage and civil partnership for all UKHLS waves and for the last 3 BHPS waves: P, Q & R (2016 2017 2018).

Note the BHPS marriage and cohabitation histories were asked in Wave 2 for the Essex sample and so will be missing for anyone who was interviewed in Wave 1 but not in Wave 2. For the BHPS regional boost samples these are asked later (waves K & L). In UKHLS, these histories were asked in Wave 1 for the GPS (General Population Sample) & EMBS (Ethnic Minority Boost sample) and then in Wave 6 for IEMBS (Immigrant and Ethnic Minority Boost sample).

Here is a list of all marriage and cohabitation history variables in the UKHLS & BHPS:


Updated by Ashley Burdett over 5 years ago

Dear Stephanie,

Thank you very much for the clarification. I think this resolves my initial concern.

Many thanks,



Updated by Ashley Burdett almost 5 years ago

Thank you very much, Alita.


Updated by Ashley Burdett almost 5 years ago

Dear Alita,

I have a couple of questions regarding "British Household Panel Survey Consolidated Marital, Cohabitation and Fertility Histories, 1991-2009" (Pronzato 2011) used in the construction of "Understanding Society: Marital and Cohabitation Histories, 1991-2015".

Specifically, I am curious about how the cohabitation spell start and end dates were constructed. Looking at the BHPS wave data, it is unclear which variables contain this information. My first thought was that mlchm and mlchy were used, however it seems that these variables refer to the date in which legal marital status changed, thus not capturing cohabitation spells. Is this correct? If so, can you please tell me which variables were used to come up marital status change dates?

Secondly, it is unclear if cohabitation spells that began and ended within a single wave, such that the individual was not in the cohabiting relationship when they completed an annual questionnaire, are captured in the BHPS wave data. Consequently, I am concerned that some cohabitation spells that lasted more than 3 months but were not ongoing during an interview are omitted, creating an inconsistency between the cohabitation spells recorded in the wave data and those recorded in the retrospective history data. Is this the case? If not, what variables provide information about intra-wave cohabitation spells?

Many thanks,



Updated by Stephanie Auty almost 5 years ago

  • Assignee changed from Ashley Burdett to Alita Nandi

Updated by Alita Nandi almost 5 years ago

  • Assignee changed from Alita Nandi to Ashley Burdett

Hi Ashley,

We used the cohabitation and marital history data collated by Pronzato from the BHPS indresp and marital and cohab files. You are correct, in the BHPS, while changes in marital status since last interview was asked changes in cohabiation spells since last interview was not asked.

Best wishes,


Updated by Understanding Society User Support Team over 3 years ago

  • Status changed from In Progress to Resolved
  • Assignee deleted (Ashley Burdett)
  • % Done changed from 70 to 100

Also available in: Atom PDF