[R] Type II and III sums of squares with Error in AOV

Γιάννης Δημάκος idimakos at upatras.gr
Wed Nov 9 17:53:43 CET 2005


On Wed, November 9, 2005 14:38, Peter Dalgaard wrote:
> Prof Brian Ripley <ripley at stats.ox.ac.uk> writes:
>
>> A multistratum aov() fit is just a list of aov() fits, so you can apply
>> functions such as Anova to the individual strata.
>>
>> However, why do you want types II and III sums of squares?  It is usual
>> to do this type of analysis only with balanced designs.  In the cases I
>> can envisage that these make any sense, they are the same as type I
>> (and in cases with only one treatment effect, they always are).
>
>
> I was about to make a similar comment. A possible exception is ANCOVA
> where you likely want to test both the within-stratum effect of a
> covariate and the effect of design factors adjusted for the covariate.

May I offer a more mundane reason for asking for Type II and PRIMARILY
Type III SS?

The editor of that particular journal who is hooked on a specific type of
software and knows only about what he/she gets from the software?

Sincerely,

Ioannis
-- 
Ioannis C. Dimakos
University of Patras
Department of Elementary Education
Patras, GR-26500 GREECE
http://www.elemedu.upatras.gr/dimakos/
http://yannishome.port5.com/




More information about the R-help mailing list