Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Phil Schumm <pschumm@uchicago.edu> |
To | Statalist Statalist <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: My ANOVA and regression results don't agree |
Date | Sat, 11 Jan 2014 12:57:36 -0600 |
On Jan 8, 2014, at 10:15 AM, David Hoaglin <dchoaglin@gmail.com> wrote: > I agree with your focus on the difference between the dummy-variable coding in the regressions and the decomposition in the ANOVA. I don't recall seeing in Jess's messages, however, that her data are balanced. > > When the data are not balanced, the contributions of the factors depend on the order in which they enter the ANOVA model, as generally happens with predictors in regression models. If Jess's are balanced, she doesn't have to face that complication in relating the results of the ANOVA to the corresponding regression results. Stata's -anova- reports partial sums of squares by default, corresponding to a regression model with all terms included. In contrast, with the -sequential- option, you're correct that the order of the terms matters in the unbalanced case, and would create another source of discrepancy between the ANOVA results and those from a regression. Even when using partial sums of squares, however, I agree that it's worth keeping in mind that the different effects are not orthogonal. -- Phil * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/