Dear all,
I am working on data including two responses (count data, relatively correlated~R2=0.4, non-normal distribution, and violated the homogeneity of covariance) and three independent variables which I am going to assess their effect on my responses. I used rankMANOVA using "ElemStatLearn" package (The rankMANOVA function calculates an ANOVA-type statistic (ATS) with (wild) bootstrap p-values for nonparametric factorial designs with multivariate data). The function also provides a possibility to run a post-hoc test like tukey. The problem I have is that I am not able to have the result for each of my responses separately like the option "summary.aov" function provides for MANOVA. Does anyone have a suggestion? Your help is really appreciated.
Hello Zohreh,
I think that the univariate comparisons you refer to are not adjusted in any way in the summary.aov function you mention. Therefore, you could just run independent rank anovas and report those results, if that helps you to understand better what's going on with your data set. I will mention that if the DVs are truly counts instead of measures, then it may well be the case that another approach to evaluating your research question(s) might be Poisson GLM, negative binomial, log-linear, among others.
Good luck with your work.
aov is designed for balanced designs, and the results can be hard to interpret without balance: beware that missing values in the response(s) will likely lose the balance. If there are two or more error strata, the methods used are statistically inefficient without balance, and it may be better to use lme in package nlme.
Balance can be checked with the replications function.
The default ‘contrasts’ in R are not orthogonal contrasts, and aov and its helper functions will work better with such contrasts.
Thank you for your help Dr. Paraman,
As you mentioned it is designed for a balanced design. My data has no missing value in the response, however, it has several zeros.
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running