Dear all; Greetings from Rwanda! I want to perform an Analysis of Variance test (ANOVA). The test for normality has shown sginficant departure from normality. The log transformation of the response does not address the isssue. I would like to know (1) what can I do so that ANOVA is applicable? (2) what can be the interpretation of result since it seems that just reporting, say: "the log of he response is significantly different from the explanatory variable level to another level" is not clear enough to most of the audience who is interested in the effect of explanatory variables' levels on the dependent variable? Thank you in advance for your suppport. Antoine Ruzindana NYIRIGIRA Institut des Sciences Agronomiques du Rwanda (ISAR) Biometric Studies Unit Rubona Research Station Planning Officer ISAR Head Quarters Kigali/Rwanda Tel: +250 (0) 788350470 You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask], leaving the subject line blank.