1 Some R Basics EPP 245/298 Statistical Analysis of Laboratory Data
October 20, 2004EPP 245 Statistical Analysis of Laboratory Data 2 Getting Data into R Many times the most direct method is to edit the data in Excel, Export as a txt file, then import to R using read.delim We will do this two ways for the energy data from Dalgaard Frequently, the data from studies I am involved in arrives in Excel
October 20, 2004EPP 245 Statistical Analysis of Laboratory Data 3 > eg <- read.delim("energy1.txt") > eg Obese Lean NA NA NA NA 8.11
October 20, 2004EPP 245 Statistical Analysis of Laboratory Data 4 > class(eg) [1] "data.frame" > t.test(eg$Obese,eg$Lean) Welch Two Sample t-test data: eg$Obese and eg$Lean t = , df = , p-value = alternative hypothesis: true difference in means is not equal to 0 95 percent confidence interval: sample estimates: mean of x mean of y > mean(eg$Obese)-mean(eg$Lean) [1] NA > mean(eg$Obese[1:9])-mean(eg$Lean) [1] >
October 20, 2004EPP 245 Statistical Analysis of Laboratory Data 5 > class(eg) [1] "data.frame" > t.test(eg$Obese,eg$Lean) Welch Two Sample t-test data: eg$Obese and eg$Lean t = , df = , p-value = alternative hypothesis: true difference in means is not equal to 0 95 percent confidence interval: sample estimates: mean of x mean of y > mean(eg$Obese)-mean(eg$Lean) [1] NA > mean(eg$Obese[1:9])-mean(eg$Lean) [1] >
October 20, 2004EPP 245 Statistical Analysis of Laboratory Data 6 > eg2 <- read.delim("energy2.txt") > eg2 expend stature Obese Obese Obese Obese Obese Obese Obese Obese Obese Lean Lean Lean Lean Lean Lean Lean Lean Lean Lean Lean Lean Lean
October 20, 2004EPP 245 Statistical Analysis of Laboratory Data 7 > class(eg2) [1] "data.frame" > t.test(eg2$expend ~ eg2$stature) Welch Two Sample t-test data: eg2$expend by eg2$stature t = , df = , p-value = alternative hypothesis: true difference in means is not equal to 0 95 percent confidence interval: sample estimates: mean in group Lean mean in group Obese > mean(eg2[eg2[,2]=="Lean",1])-mean(eg2[eg2[,2]=="Obese",1]) [1]
October 20, 2004EPP 245 Statistical Analysis of Laboratory Data 8 > mean(eg2[eg2[,2]=="Lean",1])-mean(eg2[eg2[,2]=="Obese",1]) [1] > tapply(eg2[,1],eg2[,2],mean) Lean Obese > tmp <-tapply(eg2[,1],eg2[,2],mean) > tmp Lean Obese > class(tmp) [1] "array" > dim(tmp) [1] 2 > tmp[1]-tmp[2] Lean