Follow up to Variation in responses in British Household Panel & Understanding Society surveys
Many thanks for your suggestion Jakob.
On a related matter, i.e a variation between results
in Understanding Society wave 1 and British Household
Panel wave 18.
I have run a frequency in Understanding Society wave 1 looking
at age bands, filtering on employees using:
Filter: a_jlsemp = 1 (employee)
This has produced a rather strange result when looking
at the '65 years or older' age band with 48.2% (7,107) of the responses coded as employees.
To put this into context, the percentage of people who are coded as employees in their thirties is considerably
lower, '30-34 years old' account for 3.8% (559). While '35-39 years old' account for 4.4% (653)
of the total responses received (14,759).