Support #1208

Identifying step-parent from relationship file

Added by Charlotte Edney almost 5 years ago. Updated over 4 years ago.

Data inconsistency
Start date:
% Done:




I'm using a child-centric dataset to try and get an indicator of whether the child has a step-father in the household. I'm guessing it can be done in many different ways. I've looked at the variable ypstephas which is a self-reported measure of whether the child has a step-parent living with them, however I need the step-father id to look at his characteristics. I've tried to obtain it using the relationship file however when I merge the two files together I have a large discrepancy between the children who say they have a step-parent and a step-parent being observed in the relationship file (56% of children). Do you know why this could be the case? Perhaps step-parents are not routinely interviewed?

Thanks in advance for your help.
Best regards,


Updated by Stephanie Auty almost 5 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


Updated by Stephanie Auty almost 5 years ago

  • Category set to Data inconsistency
  • Assignee changed from Stephanie Auty to Charlotte Edney
  • % Done changed from 10 to 20

Dear Charlotte,

Can you tell me which wave you are looking at? I have checked Wave 7 and am not finding the results you are reporting. You do not need the egoalt file to find the relationships as the parent pidps are included in the youth file. These are w_fnpid for natural father, w_fnspid for natural, step or adoptive father, w_mnpid for natural mother and w_mnspid for natural, step or adoptive mother. If the two are the same then the parent is a natural parent and if the pidp for natural parent is -8, inapplicable, then any pidp in the other variable is a step or adoptive parent.

Some things to be aware of: only parents in the household are recorded in this way, so if the parent now lives in a different household you would need to look back at previous waves to see if the child had that parent living with them previously. You can use w_indall for this if the child was too young to take part in the youth questionnaire at that point. Also, the young person's interpretation of w_ypstephas may not match our definition of a person living in the household. We try to interview everyone in the household (in our definition), but if that is not possible they should still be enumerated in w_indall.

Best wishes,


Updated by Charlotte Edney almost 5 years ago

Hi Stephanie,

Sorry for the delay and thanks for your response.
I was looking at a longitudinal file of all waves and a sub-sample of children from separated families (living with their mother) matched with both natural parents (where possible). I checked the fnpid vs fnspid against the ypstephas but the difference remains. Because the data is merged with parent characteristics I also looked at the partner pid and sppid of the mother and used this to see if it is different to the natural father id but again still a difference.

I've had a look at wave 7 too and generated a flag for step-parent if the fnspid and fnpid are different and I find a lot of youths who say they have a step-parent but the fnspid and fnpid don't indicate this - could you tell me how you managed to not find any differences? I think I'm missing something here.

I guess as you mentioned it's either coming from the interpretation of the variable, or they do have a step-parent but they don't live or weren't interviewed in the same household as the child. Either way it doesn't seem possible to identify characteristics of step-parents for all cases of children who report having a step-parent in the household though for some it will be possible.

Thanks for your help.
Best regards,


Updated by Alita Nandi over 4 years ago

Hello Charlotte,

You are right about this discrepancy.
In Wave 3, there were 969 who said they were living with their step parent but we don't get that info from the egoalt file, opposite case =32
In Wave 5, these numbers were 653 and 18
In Wave 7, these numbers were 807 and 17

We are looking into it to see why this may be the case. The question text reads "Do you have a step-mother or father, or someone like this, living at home with you?" So, one possible explanation is that they are interpreting their parent's partner as "someone like this"

Can you check if for these inconsistency cases, the parent of the young person is living with a partner?

Best wishes,


Updated by Charlotte Edney over 4 years ago

Hi Alita,

Thanks for your response. Your explanation definitely sounds plausible.

I've just had a look into my dataset and cross tabulated ypstephas with a new variable (partner) which combines livewith livesp and ncrr1 into one variable to indicate whether the mother has a partner (who need not be living in the same household as maybe the childs definition of "living at home with you" is not the same as a permanent member of the household). I guess these are people who don't necessarily have a partner/spouse/fnspid as they aren't interviewed. And should eliminate most of the difference. It does makes the difference a lot lower but some discrepancy still exists as you can see in the cross-tab below: 506 youths report having a step-parent but their parent does not report having a partner (from the three variables used above).

do you have |
a |
step-mother |
or father, |
or someone |
like this, |
living at | partner
home wit | 1 . | Total
yes | 1,285 506 | 1,791

Am trying to think of other possible explanations but I would imagine the interpretation of the question is the most likely cause.

Best regards,


Updated by Stephanie Auty over 4 years ago

  • Status changed from In Progress to Feedback
  • % Done changed from 20 to 80

Also available in: Atom PDF