hi, i did a binomial glmer for a repeated measure design. now i wanted to use some indicator of variance for the fixed eff. of the model for the graphical representations. sd's of the plain incident rates are huge and misleading in that context. thus i thought of extracting se's of the coefficients, back transform them and plot them as error bars. is this reasonable? any advise on how to handle this or any useful references would be greatly appreciated. greetings, kay -- View this message in context: http://n4.nabble.com/standard-errors-graphical-representation-of-glmer-tp1838338p1838338.html Sent from the R help mailing list archive at Nabble.com.