# RE: st: Creating variables of parents education when ID is known

 From "Nick Cox" To Subject RE: st: Creating variables of parents education when ID is known Date Sun, 15 Mar 2009 17:55:00 -0000

```We don't know what you don't tell us!

The FAQ remains relevant to your version of this problem. The mean of one value (or the maximum, or the minimum) is precisely that value. Section 5 is closest.

Here is another way to do it, which is not elegant, but may be easier to understand.

gen Fedu = .

qui forval i = 1/`=_N' {
su Edu if PID == FPID[`i'] & FID == FID[`i'], meanonly
replace Fedu = r(min) in `i'
}

I assume numeric variables.

Nick
n.j.cox@durham.ac.uk

Eunsu Ju, Ph.D. candidate

Thanks much, Martin.

Actually, I checked the FAQ page you posted before I sent out my question.
I think no example in the page is similar to my question. Am I missing
something?
I don't want to generate family mean but to have exact value of
someone in the family.
If there person number is constant, egen function may solve the problem.
Unfortunately, person number varies in my case.

If you think my question is answered in the page, could you please
specify example number that I pay more attention to.

On Sun, Mar 15, 2009 at 3:12 AM, Martin Weiss <Martin.Weiss1@gmx.de> wrote:

> Try http://www.stata.com/support/faqs/data/members.html

Eunsu Ju, Ph.D. candidate

>> I would like to generate a new variable which contains the information
>> of parents, e.g. dad's education.
>> My data looks like below.
>>
>>  FID  PID  Var1 Edu Var3 FFID FPID Fedu
>>
>>  1001 100  XXX  XXX XXXX   ----   ----   ????
>>  1001 200  XXX  XXX XXXX   ----   ----   ????
>>  1001 300  XXX  XXX XXXX 1001  100  ????
>>  1001 310  XXX  XXX XXXX 1001  100  ????
>>  1002 101  XXX  XXX XXXX   ----   ----   ????
>>  1002 102  XXX  XXX XXXX   ----   ----   ????
>>  1002 200  XXX  XXX XXXX 1002  101  ????
>>  1002 300  XXX  XXX XXXX 1002  101  ????
>>  1002 310  XXX  XXX XXXX 1002  200  ????
>>  1002 320  XXX  XXX XXXX 1002  200  ????
>>
>>
>> *Note: FID = Family ID; PID = Person ID; FFID = Father's family ID;
>>           FPID = Father's person ID; Fedu = Father's Education
>>
>>
>> What I want to do is to have the last (far right) column, which is not
>> included in the dataset.
>> (I want to do this kind of works for other variables, e.g. Var1 and Var3.)
>>
>> What is the best & simplist way to do this in stata?
>>
>> I think I can do this like the following.
>> 1) Split the data set into two files so that one file contains Dad's
>> FID and Dad's PID, and the other has all others.
>> 2) Then, rename FID and PID in the second file as Dad's FID and Dad's PID.
>> 3) Rename Edu (and/or other variables I want to have at Dad's level)