Dear Donald, thank you for your answer. Now I get the correct degrees. However, how is it possible that in the ANOVA between all groups only one cluster is be significant whereas in the post-hoc analysis (1 0 0 -1) there are many more? thank you, PD: I attach the 2 designs (ANOVA and post-hoc analyses) -- Maria Serra PIC (Port d'Informació Científica) Campus UAB, Edifici D E-08193 Bellaterra, Barcelona Telf. +34 93 586 8232 www.pic.es __ *Please consider the environment before printing this e-mail*