I have a large amount of NAs. Does anyone know the which command that will not throw out any rows that include NAs? I have to use gam/glm for now, but I know of other packages (RandomForests) that will use the mean of the column. Anything like this for glms? Right now this code glm2 <- glm(PA~pHNEW+SALNEW+ASPNEG+DIF232+HLSHD+SHRWV2+SLOPE+TOPON+TRI+DEM, family=binomial(link="logit"), data=predmm, *na.action=na.pass)* isn't working but this code glm2 <- glm(PA~pHNEW+SALNEW+ASPNEG+DIF232+HLSHD+SHRWV2+SLOPE+TOPON+TRI+DEM, family=binomial(link="logit"), data=predmm, *na.action=na.omit*) is. -- View this message in context: http://r.789695.n4.nabble.com/Dealing-with-NA-glm-tp3896156p3896156.html Sent from the R help mailing list archive at Nabble.com.