########################################
#
#  Script: 27 January 2011 (20110127.R)
#
########################################

# Today:
#
#   ** Analysis of Variance
#

# But, before we do that, let us look at these lines from last time.

fball <- read.csv("http://courses.kvasaheim.com/stat40x3/data/ncaa2009football.csv", header=TRUE)
boxplot(score~conference, data=fball)

model1 <- aov(score~conference, data=fball)
summary(model1)

model2 <- kruskal.test(score~conference, data=fball)
print(model2)

# Note the differences in how these are 'called' from what we have done in the past. 
# I saved the results in a variable, then called the summary function to give me a lot
# of information for the aov model and a print to get the information from the 
# Kruskal-Wallis test model.

# The standard is to have the summary() function give more information than the print() 
# function. The aov() function follows this convention. The kruskal.test() function
# violates it.

# We save the model results into variables in order to extract information from the model.

# Typical bits of information include the coefficients, which we can get from model1 by
model1$coefficients    # These are the 'effects' of each of the categories

# and the p-values, which we get from model2 by
model2$p.value

# Where do we get a list of the information we can extract? Place 1 is the help page, the 
# 'Values' section tells us. Alternatively, the names() function will also tell us
names(model1)
names(model2)

# Get into the habbit of saving the model results in a variable; it will help in the future.