Hi there, I am not an expert in this area, but as far as I know it might be in the multivariate analysis textbook by Wichern. Can't you do a dimension reduction first and then perform your MANOVA? Just a thought....I am following this question as I also have something similar in my survey data too.

Hello Isabel,

The basic design is a two-between (intensity, 2 levels; protein, 2 levels), one-within (occasion, 2 levels) manova. Between-within manovas are sometimes called "doubly multivariate."

So yes, you have three factors.

However, you may wish to consider a couple of points:

1. Do you really intend to interpret the results across arbitrary linear composite/s of the five DVs (generated by the software to maximize the variance accounted for by the between-subjects IVs), or would you prefer to address how the study factors influence scores on each of the five DVs? The first calls for the multivariate test; the second calls for univariate tests.

2. Consider using the pre-intervention score for a given dependent variable as a covariate, then use the post-intervention score as the DV score. This eliminates the within-subjects factor, and offers the ability to "adjust" for randomly occurring differences at the outset among the four treatment combinations.

3. In general, tests of change over time (e.g., meanT2 - meanT1) force you to work with difference scores that are less reliable than either of the scores from which they are derived, unless all time scores are perfectly reliable (no measurement error), which is seldom if ever the case.

Good luck with your work.

I think you can look at it as having 10 dependent variables, and then you can use techniques covered in Section 27.2 of Analysis of Messy Data Vol. 1 Designed Experiments by Milliken and Johnson.

Going back to what David said, the first question to address is: what are you looking to get out of this study? Is it to compare each of the intensity groups against each other and to compare the protein intervention groups against each other, for each of the five dependent variables separately? Or is the goal something else? It also is not clear how the study was designed: were the same subjects included for each of the intensity and protein intervention groups? All of this shapes what analyses to conduct.

Most people who think they need MANOVA really need multi-level models.

This strikes me as a two between and one within MANOVA with five dependent variables. I'd try the following.

1) Check the intercorrelations among the dependent variables at pretest and, if there are any high values, you could do a factor analysis, following a parallel analysis to determine how many factors to extract. This would be to decide how many dependent variables you really have.

2) You could then do a two between and one within MANOVA on the finallly identified dependent variables. The mechanics of doing this in SPSS are somewhat tricky.