Hi, I am looking for a way to get contrast statements (similar to the output of the TukeyHSD function) for glm fit data with a binomial distribution: glm.fit=glm(y~(treat)*(geno),family=binomial) tukey.fit=TukeyHSD(aov(glm.fit)) My y is a dataframe of number successful and number failed reproduction, the glm runs well, but not the Tukeytest - I get an error statement when I try and run a TukeyHSD on the data: Error in model.tables.aov(x, "means") : 'model.tables' is not implemented for multiple responses I am assuming this is happening because the data isn't normally distributed. If anyone could help me out that would be wonderful!! Thank you! -Adriana -- View this message in context: http://r.789695.n4.nabble.com/TukeyHSD-for-binomial-distribution-tp2401922p2401922.html Sent from the R help mailing list archive at Nabble.com.