Good evening gentlemen! I have a test in split-plot with randomized block design where my answer is a binomial variable. I wonder if there is any way I can calculate the probability of my factors considering the design errors in the case are two. I looked at various threads here and elsewhere, and unfortunately no to answer objective my problem that is very simple. My interest isn't to estimate variance components, so I see no reason to use functions like LM as found on most topics. If anyone knows a way to calculate the probability of my factors as we do in an LM model, ie, announcing the types of errors so that the probability factor in interest is not prejudiced, I'd be grateful. -- View this message in context: http://r.789695.n4.nabble.com/Split-plot-design-in-GLM-with-only-fixed-factors-tp2228126p2228126.html Sent from the R help mailing list archive at Nabble.com.