[R] aov and errors used for interaction effects

From: O.Langner <o.langner_at_psych.ru.nl>
Date: Mon, 31 Mar 2008 03:47:07 -0700 (PDT)

Dear list-members,

I just switched to R, so here probably a newbie question, I couldn't find a answer for yet:

Assume a design with three repeated measures factors (F1-F3), I tried to run the following:
aov(rt ~ F1*F2*F3 + Error(subject / (F1*F2*F3)))

Comparing the output from here and SPSS (General Linear Model -> Repeated measures) I found, that R does all calculations right for the main effects of F1-F3, but seems to do something different for the interactions: Whereas SPSS gets the F for an interaction by dividing the MS of the interaction term (e.g. F1*F2) by the corresponding MS of the interaction with subject (then F1*F2*subject), R just divides all interaction MS by the residual MS.

Could somebody point out to me, what I'm doing wrong or how I can get aov to use the appropriate error terms also for the higher order interactions?

Thanks for helping!

View this message in context: http://www.nabble.com/aov-and-errors-used-for-interaction-effects-tp16395617p16395617.html
Sent from the R help mailing list archive at Nabble.com.

R-help_at_r-project.org mailing list
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Received on Mon 31 Mar 2008 - 10:57:22 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Mon 31 Mar 2008 - 11:30:25 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive