Dear Alex, The anova() function doesn't take a type argument, producing the error. Perhaps the Anova() function in the car package will do what you want.
I hope that this helps, John > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of Hardouin Lo�c > Sent: Wednesday, March 10, 2004 10:16 AM > To: [EMAIL PROTECTED] > Subject: [R] performing type III > > Hi, > > I meet some tecnhical problem when performing type III test > in an analysis of covariance. > I code with contrast=sum and precise type="III" for the anova table. > There is a warning message like this: > Models with response "NULL" removed and type I is performed > by default. > And R-Help is not very clear on use of contrasts. > > Please don't rush in the usual holy war about this topic. And > if somebody as a concrete example... > > Thank you > > Alex > ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
