Understanding Society User Support

Many thanks for your enquiry. The Understanding Society team is looking into it and we will get back to you as soon as we can.

Best wishes,
Stephanie Auty - Understanding Society User Support Officer

Dear Vicky,

Yes, using w_mnpid and w_fnpid from the youth dataset would be an option to add youth sdq to the parent's data (or the other way around depending on how the data is set up). Alternatively, you could add youth sdq to child data first. For 10 year olds there should be both parenting style info and sdq from the youth interviews for some proportion of the sample.

Best wishes,
Stephanie Auty - Understanding Society User Support Officer

Hi Vicky,
generally, if you want to attach the child's information to the mother's record in indresp, the typical steps are

1. load a child-level data file and rename the child's pidp and all child level information to avoid overwriting of variables when merging files
2. rename _mnpidp pidp and drop those with pidp==-8 (the child's mother is not present in the household in this wave)
3. if any mother has more than one child [duplicates report _mnpid], reshape the file "wide" so that there is one row per mother (yes, you are turning that child-level data file into a mother-level data file)
4. merge with indresp using pidp.

Keeping just those with _merge==3 gives you a file at the mother-level with her child(ren)'s info attached to her own record.

The suggestion to go via the child data file is because that file is already at the child level and has a great deal of the information that you seem to be after:

1. pointers to significant others in the hh (such as _mnpid/_mnpno and _fnpid/_fnpno)
2. information about how the child is being parented: _psdqa1_dv _psdqb1_dv _psdqc1_dv _psdqa2_dv _psdqb2_dv _psdqc2_dv (for children aged 10, if the responsible adult provided the info in their adult interview). Note that there are two sets of parenting style derived variables as a child may be parented by up to two different people; pointers to them are provided in _psdq1pno and _psdq2pno. To see whether the mother or father gave the response, compare _psdq1pno and _psdq2pno to _mnpno and _fnpno, respectively.
3. and SDQ (for children aged 5 or 8, if the responsible adult provided the info in their adult interview, see _rapno4)

You can add SDQ for responding children aged 10-15 from the youth data file to the child data file using the pidp. That gives you one v large data file with any SDQ information and parenting style info collected for all enumerated children in a specific wave. There will be a lot of missing info as quite a lot of conditions must be met for there to be, e.g. a youth interview and parenting style info in a specific wave. But it gives you a complete data set from which to select the cases and variables before going through the aforementioned general steps.

Does this make things clearer?
Cheers,
Gundi