Even though I modify the syntax to "/EMMEANS=TABLES(Group*Time) COMPARE(Time)", I don't get the group-specific F results.
Reza, it might help if you posted the complete syntax rather than just the /EMMEANS sub-command. (For those who do not recognize it, this is SPSS syntax.) If you could post a reproducible example, even better. HTH.
For sure, here you are (complete SPSS syntax):
GLM F_pastB F_pastB_new s_FAT_Int s_Fat_Int_new s_FAT_att s_Fat_att_new s_FAT_sn s_Fat_SN_new s_FAT_pbc s_Fat_PBC_new F_Prisk_a F_Prisk_a_new F_Prisk_b F_Prisk_b_new s_FAT_Plan s_Fat_Plan_new BY Group
/WSFACTOR=Time 2 Polynomial
/MEASURE=Bhav Int Att Sn Pbc Risk1 Risk2 Plan
/METHOD=SSTYPE(3)
/EMMEANS=TABLES(Group*Time) COMPARE(Time)
/PRINT=DESCRIPTIVE ETASQ TEST(MMATRIX) HOMOGENEITY
/CRITERIA=ALPHA(.05)
/WSDESIGN=Time
/DESIGN=Group.
I see the problem: although SPSS shows the pairwise comparisons for each DV and within each group (p-values for t-tests), SPSS does only provides a table for the multivariate effect within each group over the two measurement points, right?
I have no straight forward solution to incorporate it into the MANOVA syntax, but since you seem to be interested in univariate post-hoc analyses, why not calculating them separately for each significant DV? A group*time ANOVA with repeated measures for the latter factor for each DV of interest?
But I would be happy if someone has a more elegant approach at hand ;-)
You bet! Rainer, this is my problem.
Then, I'll calculate them separately and multivariate F results are the same as univariate for a single DV.
Thanks indeed!
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Modeling
- Analyses of Variance