Household and individual income components
We are researching changes in household income over time, and are interested in determining which components (labour earnings, benefit income, etc.) are associated with the observed changes.
As far as I can see, the "hhresp" dataset contains an aggregate income variable but no income components. I was wondering if there are plans for those to be released?
In the meantime, I was considering using the information available in the "indresp" dataset, there are however some issues that I was hoping you could help me with.
a) w_hhnetinc1 is different from the sum of w_netinc1 for all individuals in the household. I understand this is due to the fact that w_hhnetinc1 includes additional information on proxy and individuals who did not respond to the individual questionnaire. Is that correct? Are there other possible sources of difference?
b) I also computed the sum of the individual income components for the first wave as follows:
egen a_totalinc = rowtotal(a_inc1lab a_inc2misc a_inc3prben a_inc5inv a_inc6pen a_inc7sben)
and compared it with a_netinc1. For 46794 cases the variables take the same value, but not for the other 4200 cases. I was wondering whether I am missing some income component? Or is there perhaps imputed income included in w_netinc1 but not in the component variables?
Thank you for your help,
#1 Updated by Victoria Nolan about 4 years ago
- Status changed from New to Feedback
- Assignee changed from Victoria Nolan to Alba Lanau
- % Done changed from 0 to 80
- Private changed from Yes to No
The great majority of differences are proxy cases for which we do include the component values in the data set – they are all set to zero. There is a group of cases where a_totalinc is different from a_netinc1 and there is a full interview, but this appears to be a tiny rounding error: the absolute value of the differences are all less than £0.002.
Hope this helps,
Best wishes, Victoria
On behalf of the Understanding Society Data User Support Team