Hello! I'd like to know if it is correct to test with anova two models specified like this: m1=y~x1+s(x2,by=x3),family="poisson" m0=y~x1+s(x2),family="poisson" anova(m1,m0) Cheers Anna Anna Freni Sterrantino Department of Statistics University of Bologna, Italy via Belle Arti 41, 40124 BO. [[alternative HTML version deleted]]