An ANOVA is used to assess differences on time and/or group for one continuous variable and a MANOVA is used to assess differences on time and/or group for multiple continuous variables.
In SPSS the results for ANOVA when considered only one DV did not differ when used MANOVA with multiple DV's including the one used in ANOVA. The question is, does MANOVA correct for family wise error in SPSS? As, both ANOVA and MANOVA results are alike in the situation described above.
No if I understand the question. See the text in the attached screenshot. It is available at the z-library google instructions for it's use. This a classic text in this area. Best wishes, David Booth
Hello Debanjan,
When one runs a manova in spss, the univariate tests are automatically furnished; to my knowledge, there's no syntax option to suppress them. As David Booth points out, no, the resultant probabilities reported for the anovas are not adjusted for multiple tests. It would be up to you to impose some control over aggregate type I risk by changing your criterion for statistical significance in any follow-up to a manova.
I will say that, a number of folks recommend running a discriminant analysis as a way of understanding just how your groups differ on the ensemble of DVs in your study (and they would suggest that this is more informative than a series of anovas).
Good luck with your work.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA