Project

General

Profile

Support #971

Merging Parents with SDQ

Added by Vicky Thompson almost 2 years ago. Updated over 1 year ago.

Status:
Resolved
Priority:
Normal
Category:
Data analysis
Target version:
-
Start date:
05/03/2018
Due date:
% Done:

100%

Estimated time:

Description

Hi,

I was wondering if I could get some clarification. I am wanting to merge parenting styles data with the SDQ data over waves 1-7. I have successfully merged the all of SDQ data. However, I am now wanting to add the parents. Given, that the pidp in the SDQ data is for (<16 years) and the pidp in the parenting styles data is for those 16>, could you suggest how I go about merging this information?. I had thought of using the Fpid or Mpid in some way.

Vicky

History

#1 Updated by Stephanie Auty almost 2 years ago

  • Status changed from New to In Progress
  • Assignee set to Stephanie Auty
  • % Done changed from 0 to 10
  • Private changed from Yes to No

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

#2 Updated by Stephanie Auty almost 2 years ago

  • Status changed from In Progress to Feedback
  • Assignee changed from Stephanie Auty to Vicky Thompson
  • % Done changed from 10 to 80

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

#3 Updated by Vicky Thompson almost 2 years ago

Hi Stephanie,

Thanks for your response. When merging the PSDQ with the SDQ data - the w_mnpid or w_fnpid doesnt map across as it is only present in the SDQ data and not in the PSDQ data files.Therefore I am unsure how it could be a unique identifier across those panels.

Please correct me if I have picked you up incorrectly, but you are suggesting I merge all the SDQ waves (Wave 1, 3,5 & 7) with the child data for each wave and then merge the PSDQ based on?.

Would it be possible to merge the SDQ and PSDQ data with the household identifier? I am assuming that each parent completed one PSDQ per household, per wave, rather than per child in the home. As the nature of the PSDQ is to capture the overall parenting style.

Best and Thank you in advance,

Vicky

#4 Updated by Gundi Knies almost 2 years ago

  • Category set to Data analysis
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

#5 Updated by Vicky Thompson almost 2 years ago

Hi Gundi,

Thank you so much for your reply. It is wonderfully clear and incredibly informative. It is greatly appreciated.

Best Wishes,

Vicky.

#6 Updated by Stephanie Auty over 1 year ago

  • Status changed from Feedback to Resolved
  • % Done changed from 80 to 100

Also available in: Atom PDF