Hi Andrea, Assuming that your model is something like: lm(y~x,data=mydata)
See what: cor(mydata$y,mydata$x) returns. If it is very very close to 1 or -1, there lies your problem. If one or more of your predictor variables is an almost perfect predictor of the response, you don't have much room to select different models. Jim ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see 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.