Hi Assuming you mean the MANOVA routine in SPSS (not GLM, for example), use syntax like the following to partition the interaction:
MANOVA dep BY major(1 2) place(1 3) /PRINT = SIGNIF(SINGLEDF) /CONTRAST(place) = SPECIAL(1 1 1 -2 1 1 0 -1 1). .........NOTE: if this is the contrast you want! The above will partition the 2 df (2-1)(3-1) interaction into single df contrasts. To test the simple effects (assuming same contrast) and partition simple effects into single df contrasts, use: MANOVA dep BY major(1 2) place(1 3) /PRINT = SIGNIF(SINGLEDF) /CONTRAST(place) = SPECIAL(1 1 1 -2 1 1 0 -1 1) /DESIGN major place WITHIN major(1) place WITHIN major(2). These analyses can also be done with GLM using LMATRIX command or other options. If you are using Manova in the more generic sense (i.e., multiple dependent variables), I am less certain about how the above apply. But since, multiple dep var Manova is often followed up by single dep var analyses, one could repeat above for each dep var. Hope this helps. Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax [email protected] >>> Julie Osland <[email protected]> 16-Nov-09 3:37:54 PM >>> Hi All-- I am hoping one of you can help me. Here is what I need to figure out. A student, one of our really good students, asked if you can test for interactions using MANOVA just like you do with ANOVA. I said "sure you can." So now I am in a position to have to show them how and I want to be really prepared. I found some data from a of study mine and ran a 2x3 Manova, no problem at all. But I want to be really prepared to have to answer further questions. The data has the following IVs: math major (math intensive vs. non- math intensive) and math placement (remedial, college algebra, and advanced) and what I want to know in the event I am asked, how to test for the simple main effects and Interaction Contrasts in MANOVA. I did this stuff years (and years ago) in ANOVA but have been separated from my notes and books. I've spent the better part of an hour looking at stats books and searching SPSS help to no avail. Can anyone point me to a good resource regarding how to do this in SPSS? Thanks much, Julie -- Dr. Julie A. Osland, M.A., Ph.D. Assistant Professor of Psychology Wheeling Jesuit University 316 Washington Avenue Wheeling, WV 26003 Office: (304) 243-2329 e-mail: [email protected] --- To make changes to your subscription contact: Bill Southerly ([email protected]) --- To make changes to your subscription contact: Bill Southerly ([email protected])
