I am not sure that I got a response to the question below. Could someone help?

Also, is there to run 3 way mulivariate analysis of variance (MANOVA) in SPSS V. 10? Where? ___________________________ I am a new SPSS (V.10) user. I have data on 60 Likert scale items (1-5) for 70 cases (2 groups of 35 experimental and control) with Pre and post test scores on the same instrument for both groups. I want to perform Paired or dependent sample t tests for each group on the pre and post tests. (a) on the entire scale and (b) on 6 clusters of 10 variables. Using the Analyze>Compare Means> Paired sample menu commands.... Do I compute new variables of the means of the clusters and compare those for the pre and post tests as the paired variables? (There seems to be no other way)...I am concerned that computing the means as a new variable will distort the data since SPSS will compute means on the means...is there an alternative?

Need help, Thanks Chris G. Aird Principal Sacred Heart College P. O. Box 163, San Ignacio, Cayo, Belize Phone (501)9-22758; (501)9-22102 Fax: (501)9-23759