Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

R: st: Two factor anova post test


From   "Carlo Lazzaro" <carlo.lazzaro@tin.it>
To   <statalist@hsphsun2.harvard.edu>
Subject   R: st: Two factor anova post test
Date   Tue, 19 Jul 2011 13:05:43 +0200

Dear Ricardo,
why not performing a one-way ANOVA on the significant factor (race) and, if
appropriate, perform a multiple comparison with correction?
I agree with Angel that the risk of inflating Type 1 error is probably
negligible. Anyway, you can explore in depth this issue with different
critical values.

Kindest Regards,
Carlo

-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Ricardo Ovaldia
Inviato: martedì 19 luglio 2011 12.19
A: statalist@hsphsun2.harvard.edu
Oggetto: Re: st: Two factor anova post test

Thank you Angel. The term that is significant is race, with three levels,
therefore I am not completely sure which means are different, and I do not
understand how to use -test-, or even if that is the correct way  to compare
them.

Thank you,
Ricardo.

Ricardo Ovaldia, MS
Statistician 
Oklahoma City, OK


--- On Mon, 7/18/11, Ángel Rodríguez Laso <angelrlaso@gmail.com> wrote:

> From: Ángel Rodríguez Laso <angelrlaso@gmail.com>
> Subject: Re: st: Two factor anova post test
> To: statalist@hsphsun2.harvard.edu
> Date: Monday, July 18, 2011, 4:05 PM
> Dear Ricardo;
> 
> If you were having many factors and many interactions, you
> should be
> worried about and incresed alpha error level for the
> F-test, but with
> 2 factors I doubt this is a problem. Moreover, your p level
> for the
> significance of sex is far away from the conventional 0.05,
> while the
> interaction sex*race is non-significant. Therefore you can
> conclude
> that there is a significant difference only between sexes.
> Because
> there are only two levels, there is no need to carry out
> any posthoc
> test like Scheffe, because you already know which means are
> different,
> those of men and women.
> 
> Best regards,
> 
> Angel Rodriguez-Laso
> 
> 2011/7/18 Ricardo Ovaldia <ovaldia@yahoo.com>:
> > Hello,
> >
> > I performed a two factor ANOVA; one factor (sex) has 2
> levels and the other factor (race) has 3 levels. The
> interaction term is not significant but one of the factors,
> race, is (p=0.0004). How can I determine which means are
> different while maintaining the experiment wise type I error
> rate.
> >
> > If I had only one factor, I could use -oneway- and
> then use, for example, scheffe for multiple comparisons. But
> not sure what to do with two factors.
> >
> > Thank you in advance,
> > Ricardo
> >
> > Ricardo Ovaldia, MS
> > Statistician
> > Oklahoma City, OK
> >
> >
> > *
> > *   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index