I'm having a little trouble with the glm.predict function. I read the other answers related to this question (the lm.predict question), but i'm still receiving the same warnings, then i thought i didnt got it right. Warning: " 'newdata' had 87 rows but variable(s) found have 263 rows " Commands: datamethod1 <- data.frame(method1) resultmethod1<- glm(Y ~ method1, binomial(link="probit"), data=datamethod1, control=glm.control(epsilon 1e-8, maxit = 100, trace = FALSE)) prediction1 <- predict(resultmethod1, newdata=data.frame(testgroup), type="response", se.fit=TRUE) Where 'method1' has 263 rows and 4 variables (columns) named: "minx1" "maxx1" "minx2" "maxx2" And 'testgroup' has 87 rows and the same 4 variables (columns) which have the same name. So, i hope you guys can help me figure out what i am doing wrong. thanks again Diego [[alternative HTML version deleted]]