Hi all, I am doing analyses with haplo.stats in R. I know that when doing analysis with haplo.glm, I have to define my outcome variable, i.e. if it is binomial or gaussian. When looking at interactions , eg. gender * geno or life events * geno do I have to specify whether the independent factor (eg. life events) is ordinal or binomial, ? Or does the program already know because it is a GLM anyway? Haplo.stats manual does not give any information about this. As you read the manual you assume it is not necessary to define the form of the independent variables. Does any one know? thanks, Niki -- View this message in context: http://r.789695.n4.nabble.com/defining-independent-predictors-in-haplo-stats-tp3921697p3921697.html Sent from the R help mailing list archive at Nabble.com.