Inconsistency between movest/origadd and grid references
Dear US user support team,
I am using the secure access releases of BHPS (6340) and US (6670) to track the original BHPS sample and Wales/Scotland BHPS booster sample, and have a question regarding the best way to measure inter-wave migration. I have noted from query 257 (https://www.understandingsociety.ac.uk/support/issues/257) that there are inconsistencies in how origadd captures movers in US, and would like to adopt a comparable approach for BHPS and US returns.
So far I have been using movest to measure migration for BHPS survey waves, and origadd for US survey waves. I have noted that a small proportion of non-movers (<2%) have differing grid references, and that a small proportion (<4%) of movers have identical grid references (I suppose these are intra-postcode movers, if that is the case then this proportion is rather high). These rates of 'inconsistency' are similar in BHPS and US survey waves.
Do you suggest measuring residential mobility using the survey measures, or using grid reference values (i.e. differing grid references = mover)? Presumably the latter approach will exclude intra-postcode movers?
Thanks in advance,
#1 Updated by Victoria Nolan over 3 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 - I have passed this on to the team who are looking into it for you.
Best wishes, Victoria
On behalf of the Understanding Society User Support Team
#2 Updated by Victoria Nolan over 3 years ago
- Status changed from In Progress to Feedback
- % Done changed from 10 to 80
I'm sorry for the delay in getting back to you while we have been looking into this.
On balance we would recommend using grid reference values to identify movers in a consistent way between BHPS and UKHLS. There are differences in the way the that survey information was captured in the two surveys and some limitations in the way this has been processed in UKHLS. MOVEST in BHPS is a derived variable making use of survey information on whether an interview is being carried out at a new address, combined with information on whether someone is a new member of a household. An equivalent has not yet been computed in UKHLS. This would be highly desirable and we are looking into doing this.
In the interim it should be noted that ORIGADD is a household level variable, and while it is included on the INDALL file the same value is given for all household members. It may not in fact represent the residential mobility status of all current household members if there has been household composition change.
We hope this helps,
Best wishes, Victoria.