>Hi, everyone >Previously I never did ANOVA analysis and now I am a little bit (or more) >confused with it. >Could someone help me with advise about individual groups (factor levels) >sample size in ANOVA? My data set contains 73 items, but some factor levels >contain only 3 observation. Is it justified - to conduct such an analysis? >What is the rule here? >Thank you in advance >Galina What is the maximum sample size? Having groups with n=73 and n=3 in the same analysis is likely to be very tricky - such unequal sample sizes can really compound and other potemtial problems (in particular homogeneity of variance). If n=3 all along the study will be likely underpowered but otherwise probably OK. You probably need to consider collecting more data or dropping groups from the analysis. Thom