#############################################################
## R code to reproduce statistical analysis in the textbook:
## Agresti, Franklin, Klingenberg
## Statistics: The Art & Science of Learning from Data
## 5th Edition, Pearson 2021
## Web: ArtofStat.com
## Copyright: Bernhard Klingenberg
############################################################

####################
###  Chapter 3   ###
###  Example 12  ###
####################

#############################################
##   Exploring Multivariate Relationships  ##
#############################################

# Reading in the data:
heights <- read.csv(file='https://raw.githubusercontent.com/artofstat/data/master/Chapter3/high_jump.csv')
attach(heights) # so we can refer to variable names

# Basic scatterplot
plot(x = Year, y = Winning.Height..m., pch = 16,
     col = factor(Gender),
     main = 'Winning Heights for the \n Olympic High Jump Event',
     xlab = 'Year', ylab = 'Height (m)')

legend("topleft",
       legend = levels(factor(Gender)),
       pch = 16,
       col = factor(levels(factor(Gender)))) 

# Separating observations for men and women
menObservations <- subset(heights, Gender == 'Men')
womenObservations <- subset(heights, Gender == 'Women')


# Fitting in regression model for observations for men
lmMen <- lm(Winning.Height..m. ~ Year, data = menObservations)

# Fitting in regression model for observations for men
lmWomen <- lm(Winning.Height..m. ~ Year, data = womenObservations)

# Adding the regression equations to the plot
abline(lmMen, col = 'black')
abline(lmWomen, col = 'red')

# Scatterplot using ggplot2
library(ggplot2)
ggplot(heights, 
       aes(x = Year, y = Winning.Height..m.)) +
  geom_point(aes(shape = Gender, color = Gender)) +
  geom_smooth(method=lm, se=FALSE, fullrange= TRUE,
              aes(color=Gender)) +
  labs(title = 'Winning Heights for the Olympic High Jump Event',
       x = 'Year', y = 'Height (m)') +
  theme_bw() + 
  scale_x_continuous(limits = c(1890,2030), breaks = seq(1900,2020,20))