I am working with Wave 1, and I wanted to know why the number of missing cases is so high for the mother & father educational variables (I have merged maedqf and faedqf variables from xwavedat.dta to the a_indresp file).
I know the data was collected in Waves 1 and 2, and from what I understood, all individuals in Wave 1 should have parental education information. But this seems not to be the case. Actually, the number, for example, of cases with a parental employment status when the individuals was 14 is much higher.
What does the missing refer to? Is it the case that the question was not asked? If so, do you think there may be any consequences in terms of self-selection if I use parental education in a model? Are individuals with a parental education different from those who don't have one?
Many thanks in advance!
#1 Updated by Victoria Nolan almost 4 years ago
- Status changed from New to In Progress
- Assignee changed from Victoria Nolan to Alita Nandi
- % Done changed from 0 to 10
Many thanks for your enquiry. I am assigning this to my colleague Alita who will be able to get back to you with a reply shortly.
Best wishes, Victoria
On behalf of the Understanding Society User Support Team
#2 Updated by Alita Nandi almost 4 years ago
- % Done changed from 10 to 90
In wave 1, these questions were asked of the first six months sample (a_month>=1 & a_month<=6), that is Jan-June 2009. Then in wave 2, all OSMs in the remaining monthly samples (b_month>=7 & b_month<=24) were asked these questions. In xwavedat this information was combined across the two waves. However, note that many wave 1 respondents were not interviewed in wave 2, so for them we will not have any parental education information unless they were part of the first six months sample.
You can check this by :
1. merge a_indresp with maedqf and paedqf from xwavedat.
2. Keep only those in a_indresp but with missing maedqf/paedqf information.
3. Merge these cases with b_indresp.
You will find that most of them were not interviewed in wave 2. Of those who were many were proxy respondents in both waves and so not asked these questions.
Hope this answers your question. If not, please let us know.