(Adapted from Wu, Hamada, 2009) The following experiment was performed at a pulp mill. Plant performance is based on pulp brightness as measured by a reflective meter. Each of the shift operators (dentoted A, B, C, and D) made five pulp handsheets from unbleached pulp. Reflectance was read for each of the handsheets using a brightness tester as reported in the table below:
The goal of the experiment is to determine whether there are differences between the operators in making the handsheets and reading thier brightness. a. What is the null and alternative hypothesis corresponding to the experimental goal. Calculate the means and standard deviations for each operator. Based on these summary statistics is there evidence of difference between operators? b. Use R to fit the following model to the data using the default coding contr. treatment (4)
wherei= 1...,4,j= 1.....5 , where yij is the jfh observation with operator i. ●Interpret the model parameters η, Ti, Eij and their associated p-values. ●What do you assume about Eij? Check if these assumptions are fuflilled.
sapply(split(reflect,operator),mean)
sapply(split(reflect,operator),sd)
contrasts(pulpdat$operator) <- contr.treatment(4)
contrasts(pulpdat$operator) <- contr.sum(4)
pairwise.t.test(reflect,ope
# Question 1
# Read in data for question #1
pulpdat <- read.csv("/Users/nathantaback/Dropbox/Docs/sta305/2015/assignments/Assignment3/pulpdat.csv")
attach(pulpdat)
#Calculate means and sd for each operator
sapply(split(reflect,operator),mean)
sapply(split(reflect,operator),sd)
# Model using treatment contrasts
contrasts(pulpdat$operator) <- contr.treatment(4)
mod1 <- lm(reflect~operator,data=pulpdat)
summary(mod1)
# Model using deviation coding
contrasts(pulpdat$operator) <- contr.sum(4)
mod2 <- lm(reflect~operator,data=pulpdat)
summary(mod2)
#
pairwise.t.test(reflect,operator,conf.level=0.95,p.adj = "bonf")
TukeyHSD(aov(reflect~operator),conf.level=0.95)
# Question 3
# Read in BHH Data from chapter 5, problem 6
prb0506 <- read.table("~/Dropbox/Docs/sta305/BHHData/BHH2-Data/prb0506.dat", header=TRUE, q
原创声明:本文系作者授权腾讯云开发者社区发表,未经许可,不得转载。
如有侵权,请联系 cloudcommunity@tencent.com 删除。
原创声明:本文系作者授权腾讯云开发者社区发表,未经许可,不得转载。
如有侵权,请联系 cloudcommunity@tencent.com 删除。