2013-03-08_REVISION - C1981008 (2)

Home Browse Search

R code: (R is free statistical software, which can solve the problems that SAS can. I think it is better to use R instead of SAS in this case, since you can download R and try the code by yourself. I attach the two files "soil_first.txt" and "soil_second,txt" in the email, by which you can try the following Rcode.) 1. Code for the first data set: library(MASS) s o il_frs t <- read.table ( "so il_fi rs t.txt ") names (soil_first)< -c( "soil ", "type ") sub‹- soil_first$soil [ soil _first$type== "Substitutesoil "l barxp < -so i1_firs t$soil [soil_ftrs t$type = = "Soi IBarxprofite "] library(MASS) # the Wilcoxon rank sum test between "Substitute Soil" and "Soil Barxprofile" w ilc ox. t es t(s u b, b arxp ) #Wilcoxon rank sum test with continuity correction #data: sub and barxp #W = 55, p -value = 0.2036 #alternative hypothesis: true location shift is not equal to 0 #Conclusion: there is no significant difference between the distributions for "Substitute Soil" and "Soil Barxprofile" # the Wilcoxon rank sum test between "Substitute Soil" and "BarxLiftA" barxa <- soil_first$soil [so il_first$type == "BarxLiftA"] w ilco x. test(s ub,barxa) #Wilcoxon rank sum test with continuity correction Hata: sub and barxa #W = 29.5, p -value = 0.01641 #alternative hypothesis: true location shift is not equal to 0 #Conclusion: there is significant difference between the distributions for "Substitute Soil" and " BarxLiftA" # the Wilcoxon rank sum test between "Substitute Soil" and "BarxLiftB" barxb <- soil_first$soil [ soil_first$type== 'BarxLiftB"] wilcox.test(sub,barxb) #Wilcoxon rank sum test with continuity correction #data: sub and barxb #W = 25.5, p -value = 0.7648 #alternative hypothesis: true location shift is not equal to 0 #Conclusion: there is no significant difference between the distributions for "Substitute Soil" and " BarxLiftB" 2. Code for the second data set: library(MASS) PLTF 002496