2. The article “Origin of Precambrian Iron Formations” (Econ Geology, 1964: 1025

Question

2. The article “Origin of Precambrian Iron Formations” (Econ Geology, 1964: 1025- 1057) includes data on total Fe of four types of iron formation. The data is available from as Fe.csv. Use Rcmdr to analyze the data.

A. Give the sample mean and standard deviation for each formation type. (1 pt)

B. Create side-by-side boxplots of Fe by formation type. (1 pt)

C. Construct a side-by-side qqplot for each formation type. (2 pt)

D. Construct a scatter plot the residuals versus the fitted values. (2 pt)

E. Do the assumptions for the one-way ANOVA appear to be met? Discuss. (2 pt)

F. Carryout the (one-way) ANOVA analysis. Give the ANOVA table and conclusion. (2pt)

G. Give the Tukey confidence intervals for the pairwise differences. (1 pt)

H. Based on the Tukey confidence interval, can we conclude that there is a difference

between mean Fe for Magnetite versus Carbonate formations? Explain. (1 pt)

Explanation / Answer

#Data -> Import Data -> from text file -> Navigate to fe.csv and click OK
fe <- read.csv("C:\Users\Aswin\Downloads\fe.csv", header = T, sep = ",")

#a)Statistics -> Summaries -> Numerical Summaries(select summarize by groups and
#select Formation. Go to statistics tab and check mean and standard deviation

numSummary(fe[,"Fe"], groups=fe$Formation, statistics=c("mean", "sd"),
quantiles=c(0,.25,.5,.75,1))

#b)Graphs -> Boxplot (select plot by group and click on Formation.)
Boxplot(Fe~Formation, data=fe, id.method="y")

par(mfrow=c(1,4))

with(fe, qqPlot(Fe[Formation == "Carbonate"], dist="norm", id.method="y", id.n=2,
labels=rownames(fe)))
with(fe, qqPlot(Fe[Formation == "Silicate"], dist="norm", id.method="y", id.n=2,
labels=rownames(fe)))
with(fe, qqPlot(Fe[Formation == "Magnetite"], dist="norm", id.method="y", id.n=2,
labels=rownames(fe)))
with(fe, qqPlot(Fe[Formation == "Hematite"], dist="norm", id.method="y", id.n=2,
labels=rownames(fe)))

#Select simple linear model from Models tab
LinearModel.1 <- lm(Fe ~ Formation, data=fe)

#d)Select Basic diagnostic graphs from graphs in Models tab
oldpar <- par(oma=c(0,0,3,0), mfrow=c(2,2))
plot(LinearModel.1)
par(oldpar)

Related Questions

Navigate

• Browse (All)
• Browse 2
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.