Hi, I've got this model and following Hothorn et al advices, I used glht for a post hoc comparison
modezqM<-glm(rojos~estacion*zona3,quasipoisson,subset=(edadysexo=="M")) anova(modezqM,test="F")
Df Deviance Resid. Df Resid. Dev F Pr(>F) NULL 293 41148 estacion 1 3647.0 292 37501 23.6250 1.927e-06 *** zona3 2 931.4 290 36569 3.0167 0.050510. estacion:zona3 2 2421.7 288 34148 7.8438 0.000482 ***
summary(glht(modezqM,linfct=mcp(zona3="Tukey",interaction_average=T)))
Estimate Std. Error z-value Pr(>|z|) norte - turist == 0 -0.05182 0.23701 -0.219 0.9727 occid - turist == 0 -1.05970 0.41618 -2.546 0.0271 * occid - norte == 0 -1.00788 0.45830 -2.199 0.0665 . I'm trying to understanding the way this command made pairwise comparisons, but, where does this z-value comes from? I'm suspected this not come from Tukey's test, but i don't know it's Wald or something i've never use I know it is not an habitual doubt. Sorry and thanks in advance I suspect it ----- Mario Garrido Escudero PhD student Dpto. de Biolog?a Animal, Ecolog?a, Parasitolog?a, Edafolog?a y Qca. Agr?cola Universidad de Salamanca -- View this message in context: http://r.789695.n4.nabble.com/pairwise-comparisons-with-multcomp-package-tp4384499p4384499.html Sent from the R help mailing list archive at Nabble.com.