Re: st: Using pathreg in STATA

["Michael N. Mitchell" <Michael.Norman.Mitchell@gmail.com>]

Dear Amal

  Indeed, I think you answered your own question. As the author, Phil Ender, noted... in 
many ways this is just a regression model, so you can use the same principles you would 
have used from regression. This includes, as you suggest, including an arrow from age to 
math, and an arrow from age to science. In that case, the effects with respect to math 
will be adjusted for age, and the effects with respect to science will be adjusted for age.

Best regards,

Michael N. Mitchell
Data Management Using Stata      - http://www.stata.com/bookstore/dmus.html
A Visual Guide to Stata Graphics - http://www.stata.com/bookstore/vgsg.html
Stata tidbit of the week         - http://www.MichaelNormanMitchell.com



On 2010-09-07 9.29 AM, Amal Khanolkar wrote:
> Dear Michael,
>
> Thank you for your reply.
>
> If I understand right, in the example that you sent me, read and write
> are corelated but their effects are not estimated as such. My problem
> is, say hypothetically, that we had an arrow with age pointing to math
> and another similar 'age-arrow' pointing to science. I know that age has
> an effect on both math and science. Age is not in the pathway under
> study but has to be adjusted for.
>
> Does one then just include another (math age)&  (science age) command to
> signify these relationships in the pathreg command (thereby adjusting
> for age aswell)??
>
> Thanks,
>
> /Amal.
>
> Amal Khanolkar,
> PhD Student,
> CHESS | Centre for Health Equity Studies,
> Karolinska Institutet/Stockholm University,
> Sveavägen 160, Sveaplan - 10691 Stockholm.
>
> Phone: +46 (0)8 162584
> www.chess.su.se
>
> ----- Original Message -----
> From: "Michael N. Mitchell"<Michael.Norman.Mitchell@gmail.com>
> Date: Monday, September 6, 2010 8:38 pm
> Subject: Re: st: Using pathreg in STATA
> To: statalist@hsphsun2.harvard.edu
>
> > Dear Amal
> >
> >    If I understand your question correctly, the example at
> >
> > http://www.ats.ucla.edu/stat/stata/faq/pathreg.htm
> >
> >    would be such an example. In that example, read is adjusted for
> > write, and write is
> > adjusted for read.
> >
> > I hope that helps,
> >
> > Michael N. Mitchell
> > Data Management Using Stata      -
> > http://www.stata.com/bookstore/dmus.htmlA Visual Guide to Stata
> > Graphics - http://www.stata.com/bookstore/vgsg.html
> > Stata tidbit of the week         -
> > http://www.MichaelNormanMitchell.com
> >
> >
> > On 2010-09-06 8.34 AM, Amal Khanolkar wrote:
> > > Hi everyone,
> > >
> > > I have a question regarding the 'pathreg' command. Is it possible to
> > > adjust for covariates/confounders in addition to just studying
> > effects>  between variables in the path model? If so, could anyone
> > please give an
> > > example of a command with adjustment for confounders using pathreg.
> > >
> > > Thanks,
> > >
> > > /Amal.
> > >
> > > Amal Khanolkar,
> > > PhD Student,
> > > CHESS | Centre for Health Equity Studies,
> > > Karolinska Institutet/Stockholm University,
> > > Sveavägen 160, Sveaplan - 10691 Stockholm.
> > >
> > > Phone: +46 (0)8 162584
> > > www.chess.su.se
> > >
> > > *
> > > *   For searches and help try:
> > > *   http://www.stata.com/help.cgi?search
> > > *   http://www.stata.com/support/statalist/faq
> > > *   http://www.ats.ucla.edu/stat/stata/
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/