[R] ANCOVA with defined error terms

hpdutra hpdutra at yahoo.com
Sat Aug 22 17:36:35 CEST 2009


Thank you Richard, this works. But, the model you suggested me lacks some
between subjects  interactions, namely:
 veget:time:block
 fruit:time:block. 
 According to Sokal and Rohlf I need to report those as well. 
Also, any ideas why the sum of squares on my model are different from yours

summary(mymodel<-(aov(mice~veget*fruit*time+(time*block)+(time*block*veget)+(time*block*fruit)+coon+Error(block/plot,
data = track))))
reports veg:fruit with a sum squares of 80.5

while yours 
summary <- (ricsmodel<-aov(mice ~ coon + block*veget*fruit*time -
block:veget:fruit:time+ Error(block/plot), data = track))
gives me a veget:fruit sum of squares of 253.2
the residuals degrees of freedom also differ my model gives 4 DF for the
between subjects while yourr does not report one for the between subjects. 
I know I'm asking too much, but I'm just trying to understand.
Thanks
Beto 
s 



Richard M. Heiberger wrote:
> 
> Yes, I meant summary().  anova() isn't defined for aovlist objects and 
> summary() is.
>> Warning message:
>> In aov(kotz.mice ~ kotz.coon + block * veget * fruit * time -
>> block:veget:fruit:time +  :
>>   Error() model is singular
>>   
> 
> You will need to investigate the singular Error() model.  You might want
> to use the simpler Error model
> 
> Error(block)
> 
> My guess is the plot term is redundant.
> 
> What are kotz.mice?  That variable isn't in the model you showed.
> 
> For more detailed help you will need to
> "provide commented, minimal, self-contained, reproducible code."
> 
> 
> Rich
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 
> 

-- 
View this message in context: http://www.nabble.com/ANCOVA-with-defined-error-terms-tp25055311p25094846.html
Sent from the R help mailing list archive at Nabble.com.




More information about the R-help mailing list