When we ran a regular ANOVA we showed that both mice and raccoons respond positively to cover. So we decided to use raccoons as a co-variate because those are mouse predators and their presence per se could explain part of the variation on mice activity. That is the reason why I'm running this ANCOVA. Thus your model that places coon sequentially first makes more sense for this question.
"Error() model is singular" Some authors claim that one should test for the interactions of block versus treatment, to identify any spatial correlation with the data. That is the reason why I tried to incorporate the interactions of factors with block in my model. But like you said I don't have enough degrees of freedom for it. Is there an R function that calculates the degrees of freedom for complex designs? I did not understand what you meant by "Error() notation. Numerators and Denominators for F tests are specified indirectly by the relation of the linear subspaces in the treatment model and the Error model." If this is last question is outside the scope of R-help, please just give me a reference that covers it and I'll do my homework. Thank you very much Richard, this was wonderful help. -- View this message in context: http://www.nabble.com/ANCOVA-with-defined-error-terms-tp25055311p25110724.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@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.
