# Exercise 6.3 # Statistical Methods in Biology: Design and Analysis of Experiments and Regression # by S.J. Welham, S.A. Gezan, S.J. Clark & A. Mead (2014) # Chapman & Hall/CRC Press, Boca Raton, Florida. ISBN: 978-1-4398-0878-8 # Data from Rothamsted Research (A. Karp) # Version 1, 13/08/2014 # Set working directory - use setwd() function or from Session menu in RStudio # e.g. setwd("d:/stats4biol/data) # load external packages - availabel from CRAN library(ggplot2) library(lsmeans) # tables of predicted means & contrasts # Read data & assign factors scab <- read.table('scab.dat', sep="", header=TRUE) scab$Treatment <- as.factor(scab$Treatment) scab$Row <- as.factor(scab$Row) scab$Col <- as.factor(scab$Col) scab$Sulphur <- as.factor(scab$Sulphur) scab$Timing <- as.factor(scab$Timing) summary(scab) # One-way ANOVA of raw data scab.aov <- aov(Scab ~ Treatment, data=scab) summary(scab.aov) # Residual plots plot(scab.aov, ask=FALSE) hist(residuals(scab.aov)) # Take logit-transformation scab$logit <- log(scab$Scab/(100-scab$Scab)) summary(scab) # One-way ANOVA of logit-transformed data scab.aov.2 <- aov(logit ~ Treatment, data=scab) summary(scab.aov.2) # Residual plots plot(scab.aov.2, ask=FALSE) hist(residuals(scab.aov.2)) # Get table of predcited means with back-transform scab.lsm <- lsmeans(scab.aov.2, ~Treatment) scab.lsm.df <- summary(scab.lsm) scab.lsm.df$bt <- 100*exp(scab.lsm.df$lsmean)/(1+exp(scab.lsm.df$lsmean)) scab.lsm.df # End of file