Dear readers,

Using SPSS MANOVA I got the following for me inexplicable results:

With the syntax:

MANOVA y BY a(1 2) typevide(1 3) typeargu(1 2) /METHOD=UNIQUE /ERROR WITHIN + residual /DESIGN a c a BY c b BY c b WITHIN a (1) b WITHIN a (2) a BY b BY c.

I got (amongst others)

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Tests of Significance for Y using UNIQUE sums of squares Source of Variation SS DF MS F Sig of F

WITHIN+RESIDUAL 76,45 107 ,71 B WITHIN A(1) 12,76 2 6,38 8,93 ,000 B WITHIN A(1) 53,85 2 26,93 37,69 ,000 (Total) 165,62 118 1,40 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

(of course, the residuals term is zero).

With the following syntax:

MANOVA y BY a(1 2) b(1 3) c(1 2) /METHOD=UNIQUE /ERROR WITHIN /DESIGN b WITHIN a (1) b WITHIN a (2).

which should give - I assume - the same results for the reported effects, I got

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Tests of Significance for Y using UNIQUE sums of squares Source of Variation SS DF MS F Sig of F

WITHIN CELLS 76,45 107 ,71 B WITHIN A(1) 11,87 2 5,94 8,31 ,000 B WITHIN A(2) 52,68 2 26,34 36,87 ,000 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Can anybody tell me where the difference is SSs comes from? Am I missing something here.

Thanking you in advance,

Ivo van der Lans

-- Ivo van der Lans