Another addition:
If you want to know if the difference of means of a continuous
variable between two groups depends on a third variable
(neighborhood), wouldn´t it make sense to run a regression with an
interaction term group*neighborhood?
Best regards,
Angel Rodriguez-Laso
2009/5/18 Verkuilen, Jay <[email protected]>:
> In addition to what everyone else has already said, it sounds like you
> potentially have a multivariate problem here, as you have several Y
> variables you want to compare two groups over. This suggests to me that
> several univariate ANOVAS (or dummy variable regressions, the same
> thing) might be profitably replaced with -hotelling-, -manova- or
> -discrim-. These are all essentially variations on the same thing. If
> not replaced, it makes sense to preface univariate analyses with a
> multivariate one.
>
> JV
>
> *
> * 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/
