Re: st: How to create a new variable which contains the responsefrom anot=

[Klaus Haberkern <haberkern@soziologie.unizh.ch>]

Hi Chiungya,

if you have no more than two members in a group, the following syntax 
may work
with "[_n+1]" you can refers to the case in the following line


It's important that the data are sorted as in your example:

Group    member        V1
=E3=80=801       1            3.2
=E3=80=801       2            4.5
=E3=80=802       1            3.1
=E3=80=802       2            3.3
=E3=80=803       1             .
=E3=80=803       2            2.3




gen v2=.

*[_n+1] taking the values from the next line (that is person in the 
group) in the data, [by group, sort] only when line below represents 
person in the same group
by group, sort: replace v2=v1[_n+1] if v2==.

*[_n-1] taking the values from the line before in the data, [by group] 
only when line above represents a person of the same group
by group: replace v2=v1[_n-1] if v2==.


If you have more persons in a group you have to make sure, that you 
refer to the right case in a group.
For example, if you analyse families, and you want to relate both 
parents, but not the children in a household, you have to know, who is 
the partner of the respondent: e.g. vars id (own id) and pid (id of 
partner)

sort group id
by group: replace v2=v1[_n+1] if v2==. & pid==id[_n+1]
by group, sort: replace v2=v1[_n-1] if v2==. & pid==id[_n-1]

if difference between id and pid >1 or <-1 you have to gone on like that

by group: replace v2=v1[_n+2] if v2==. & pid==id[_n+2]
etc

Hope this helps,
Klaus

______

_________________________
Klaus Haberkern
Soziologisches Institut
Universität Zürich
Andreasstrasse 15
CH-8050 Zürich

Tel.: ++41-(0)44-635 23 43
E-Mail: haberkern@soziologie.unizh.ch
Internet: www.suz.unizh.ch/ages



Chiung-Ya Tang schrieb:

> her member
> Date: Sun, 4 Mar 2007 20:06:50 -0500
> Message-ID: <002301c75ec2$8ebe21c0$6401a8c0@Saturdaynight>
> MIME-Version: 1.0
> Content-Type: text/plain;
>   =09charset=3D"UTF-8"
> X-Mailer: Microsoft Office Outlook 11
> X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3028
> Thread-Index: Acdewo3xDbzlOHAwQ/aMhHOZCSj6qg=3D=3D
> Content-Transfer-Encoding: 8bit
> X-MIME-Autoconverted: from quoted-printable to 8bit by hsphsun2.harvard.edu=
>   id l2517LJ10855
> Sender: owner-statalist@hsphsun2.harvard.edu
> Precedence: bulk
> Reply-To: statalist@hsphsun2.harvard.edu
> Errors-To: statalist-owner@hsphsun2.harvard.edu
> Hi,
>
> I am learning STATA and would like to use it to analyze a set of dyadic dat=
> a. The data is as following:
>
> Group    member        V1
> =E3=80=801       1            3.2
> =E3=80=801       2            4.5
> =E3=80=802       1            3.1
> =E3=80=802       2            3.3
> =E3=80=803       1             .
> =E3=80=803       2            2.3
>
> I wish to generate a new variable--V2 in which responses from V1 are exchan=
> ged within a group. So, it should look like this:
>
> Group         member          V1      V2
> =E3=80=801              1            3.2     4.5
> =E3=80=801              2            4.5     3.2
> =E3=80=802              1            3.1     3.3
> =E3=80=802              2            3.3     3.1
> =E3=80=803              1             .      2.3
> =E3=80=803              2            2.3      .
>
>
> I searched commands as well information from Statalist archives but still h=
> aven't figured out how to do that correctly. While keeping learning new com=
> mands, I definitely appreciate any help from you. Thank you.
>
> Sincerely,
> Chiungya
> =E3=80=80
> =E3=80=80
> =E3=80=80
> =E3=80=80
>
>
> *
> *   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/
>
>   
*
*   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/