Elizabeth Beck
2013-Jul-10 21:28 UTC
[R] permanova for multivariate repeat measures toxicology data set
Hello, I would like to use a permanova to analyze my repeated measures, multivariate data set using the package vegan (adonis function). I have several explanatory variables and many clinical biochemistry/hematological response variables (all continuous, non-normally distributed). My explanatory variables are: ID - 29 levels - individual subjects Treatment - 2 levels Sex - 2 levels Time - 11 levels - subjects were treated 11 times I am interested in (1) whether the response variable changes over time with repeated exposure and (2) whether the 2 treatment groups are significantly different from each other and whether this effect varies with time. I can run the adonis function from package vegan but am unsure how to incorporate the repeated measures on the same subjects into the analysis. adonis(hemvar~ hemat$Treatment * hemat$Exposure * hemat$SEX * hemat$ID, permutations=999, method="euclidean") Any help would be greatly appreciated! An example of one of my data sets....> str(hemat)'data.frame': 87 obs. of 14 variables: $ ID : Factor w/ 29 levels "G-BB","G-BW",..: 1 1 1 2 2 2 3 3 3 4 ... $ Time : Factor w/ 3 levels "Tx4","Tx5","Tx6": 1 2 3 1 2 3 1 2 3 1 ... $ Treatment: Factor w/ 2 levels "Control","Experimental": 1 1 1 2 2 2 1 1 1 1 ... $ SEX : Factor w/ 2 levels "Female","Male": 2 2 2 2 2 2 1 1 1 2 ... $ PCV : int 45 33 38 50 49 53 45 37 39 36 ... $ WBC : num 10.3 6.8 11.6 19.1 9.7 9.8 20 34.3 33.4 20.7 ... $ BASO : num 0.206 0.068 0.464 0.382 0.388 ... [[alternative HTML version deleted]]