BHPS: analysing results from children who have left the household and matching them to their parents
Hi, I'm hoping someone could offer some guidance.
I'm currently working with the BHPS and trying to see the effect of trade union membership on children's future membership.
I'm finding it very difficult to be able to match data to allow me to do this.
I am able to match the data in a single wave using the xFPID number.
However, I have no idea how to do this for future waves and for children who are no longer in the household.
It would be great if anyone could offer some guidance
#1 Updated by Gundi Knies about 4 years ago
- Category set to Data analysis
- Status changed from New to In Progress
- Assignee set to marianne walker
- Priority changed from High to Normal
- Target version set to BHPS
- % Done changed from 0 to 70
there are different ways of doing this.
One option is to start off with the data file XWAVEDAT. It includes every person who has ever been enumerated in any wave of the BHPS, and in principle you only need the PID from this data file. Next add the FPID from each wINDALL to every person's record on XWAVEDAT by matching on PID.
Note that there may be data inconsistencies, i.e. the FPID may vary over time from 0 when the father is not co-resident to a positive number when the father is co-resident but also to a different positive number if a different co-resident person is identified as the father in one of the waves. You will need to clean this, assuming you just want to use the information relating to 1 biological father. When you have decided the FPID for each person, rename the PID (e.g., to KPID for Kid PID) and FPID to PID. Then merge this file to the wave specific wINDRESP files using PID to add the union status of the father. After that you will need to rename the variables relating to the father and undo the renames of PID to KPID and FPID to PID again. Now that the PID refer to the child again, you can merge on PID to add the child’s characteristics from other files.
Hope this helps,