Ancova repeated measures or manova rm.my participants receive one of the interventions.my other question is if the multivariate tests table in anova rm did not show any significant value but within and between tables had significant esults can we ignore the multivariate table and report the result of within and between?
Jos Feys' is correct but there are a couple of potential problems that you did not mention. First, how many participants are in your study and how were they selected. Second this is not a hunt for all p-values< 0.05. I would not recommend forgetting all p-values you don't like and reporting only the ones that you do. Further there are requirements that must be met before you can trust your p-values. All of this and a few more things are necessary such as graphs and so on. That's why I tend to recommend the equivalent RM multivariate regression which I believe is easier to interpret than the usual manova approach. I would recommend that you consult a standard multivariate analysis text that includes both approaches and choose the easier for you. I would suggest the text by Johnson and Wichern available at the z-library, google this for instructions on use. Further there are helpful analysis programs available there or on the web for doing these analyses correctly. I recommend checking these for your favorite stat package and videos on their use. May the Force be with us all. David Booth
What is the purpose of the 3 repeated measures? Are they taken at different times, and are you expecting to see change over time, or perhaps a group x time interaction? Or is there some other reason? Thanks for clarifying.
You have told us your data, but not your research question. Anyone telling you what to do right now is not helping you, because if you run some sort of statistical model you will have an answer to a question, but it may not answer your question!
So what's the study question!
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA