--- Maarten buis <[email protected]> wrote:
> I am still puzzled:
>
> Do you have mothersnum in both datasets or just in
> the dataset of the
> children?
I have it in both datasets. But I only need it for the
childrens dataset. As for the mothers I'm only using
their roster number.
>
> Do you have in both datasets a variable that
> idetifies to which familiy
> someone belongs?
I have the houshold number (hhnum), which is a
different value for each family. I also have the
roster number which is basically a different number
for each hh member. The mothernumber indicates the
roster number of the mother.
>
> Is there in both datasets any information available
> that alows you to
> see if two observations belong together?
>
> Well, I know for each child who is the mother by
looking at the mothersnum which is the roster number
of the mother. I also know the family they belong to.
>
> --- Ana Gabriela Guerrero Serdan
> <[email protected]> wrote:
>
> > Sorry Maarten but I didnt explain myself
> correctly.
> >
> > If I would have a unique identifier to match
> children
> > and mothers I would use it. What I have is a
> variable
> > for each children called mothersnum which tells me
> the
> > roster num of the mother. For mothers there is no
> > indication of what is the roster number of the
> child.
> > So if I merge mothersnum in children database with
> var
> > mothersnum in full database I would get nothing
> since
> > the mothersnum in the full database is "." If I
> use
> > roster number I would then only match children
> with
> > themselves.
> >
> > Apologies if this was a confusing query but here
> is
> > what I have worked out and seems to be working
> now.
> >
> > I have created a new identifier (id).
> >
> > In full database:
> >
> > gen id=rosternum
> > gen msex=sex */doesnt necessarily reflects only
> mother
> > but just to differentiate from the sex var of
> children
> > with is also sex*/
> >
> > gen mage=age */doesnt necessarily reflects only
> mother
> > but just to differentiate from the age var of
> children
> > with is also age*/
> >
> > gen edumother=education */doesnt necessarily
> reflects
> > only mother but just to differentiate from the
> > education var of children with is also education*/
>
> >
> > sort hhnum id
> > save as "mothersdatabase"
> >
> > list hhnum roster id mothersnum mage msex
> edumother
> >
> >
>
+---------------------------------------------------------+
> > hhnum roster id mother~m mage msex
> > edumot~r
> >
>
---------------------------------------------------------
> > 1. 1 1 1 . 62 1
> 1
> >
> > 2. 1 2 2 . 62 2
> 1
> >
> > 3. 1 3 3 . 41 1
> 1
> >
> > 4. 1 4 4 . 28 2
> 1
> >
> > 5. 1 5 5 . 10 1
> 1
> >
> >
>
---------------------------------------------------------
> > ............
> >
>
---------------------------------------------------------
> > 11. 1 11 11 . 6 2
>
> > 1
> > 12. 1 12 12 9 0 1
>
> > 2
> > 13. 1 13 13 . 38 1
>
> > 1
> > 14. 1 14 14 . 32 2
>
> > 1
> > 15. 1 15 15 . 10 1
>
> > 1
> >
>
---------------------------------------------------------
> > ............
> >
>
---------------------------------------------------------
> > 21. 1 21 21 18 3 2
>
> > 2
> > ..........
> >
> > In childrens database:
> > gen id=mothersnum
> > sort hhnum id
> > save
> > merge hhnum id using "mothersdatabase", nokeep
> >
> > list AI01 HR01 id sex age mothernum msex mage
> > edumother
> >
> >
> >
>
+------------------------------------------------------------------+
> > AI01 HR01 id sex age mot~rnum msex mage
> > edumot~r
> >
> >
>
------------------------------------------------------------------
> > 1.1 12 9 1 0 9 2 30
>
> > 1
> > 2.1 21 18 2 3 18 2 29
>
> > 1
> > 3.1 22 18 2 0 18 2 29
>
> > 1
> > 4.2 6 4 2 1 4 2 25
>
> > 1
> > 5.2 7 4 2 1 4 2 25
>
> > 1
> >
> >
>
------------------------------------------------------------------
> > .........
> >
> >
> > So for child with rosternum 9 the mothers msex=2
> > (female ofcourse!) and mage=30.
> >
> > Again apologies for my confusion.
> >
> > regards,
> > Gaby
> >
> >
> >
> >
> >
> >
> >
> > --- Maarten buis <[email protected]> wrote:
> >
> > > If you would manually match a child to a mother,
> > > what variables would
> > > you use?
> > >
> > > --- Ana Gabriela Guerrero Serdan
> > > <[email protected]> wrote:
> > >
> > > > Dear Maarten,
> > > >
> > > > Yes, ofcourse. The problem is that if I merge
> by
> > > using
> > > > hhnumber and mothers number there is a
> mismatch
> > > > because it means i merge the children dataset
> with
> > > its
> > > > own number as it is indicated in the
> > > mothersnumber. I
> > > > cant use roster number neither.
> > > >
>
=== message truncated ===
Gaby Guerrero Serdan
Deparment of Economics
Royal Holloway, University of London
TW20 OEX
Egham, Surrey
England, UK
http://www.rhul.ac.uk/economics/About-Us/postgrads.html
http://www.flickr.com/photos/49939890@N00/
Tel: +44 7912657259
____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/