R Workshop: Practical DAY 2
dr. Annelies Agten
2026-06-09
This document contains the solutions to the exercises from the practical session of Day 2 of the Introduction to R workshop. The session focuses on data exploration, both numerical as well as graphical. The exercises in the document are all based on base R. However, the ggplot options are also provided. Note that you can enhance any figure as much as you personally like. There is no ‘right’ or ‘wrong’ or only one valid solution.
Before you start:
- Set working directory
- Read in iris dataset (‘iris.csv’ file)
- Read in hair and eye color dataset (‘HairEyeColor.csv’)
- Check your datasets
# Note that the path to your folder will be different
setwd("~/ADS Courses/Intro to R/DAY 2")
# Import datasets HairEyeColor and Iris
haireye <- read.csv("HairEyeColor.csv", header=T)
str(haireye)
#> 'data.frame': 592 obs. of 2 variables:
#> $ hair: chr "Black" "Black" "Black" "Black" ...
#> $ eye : chr "Brown" "Brown" "Brown" "Brown" ...
View(haireye)
iris <- read.csv("iris.csv", header=T)
str(iris)
#> 'data.frame': 150 obs. of 5 variables:
#> $ Sepal.Length: num 5.1 4.9 4.7 4.6 5 5.4 4.6 5 4.4 4.9 ...
#> $ Sepal.Width : num 3.5 3 3.2 3.1 3.6 3.9 3.4 3.4 2.9 3.1 ...
#> $ Petal.Length: num 1.4 1.4 1.3 1.5 1.4 1.7 1.4 1.5 1.4 1.5 ...
#> $ Petal.Width : num 0.2 0.2 0.2 0.2 0.2 0.4 0.3 0.2 0.2 0.1 ...
#> $ Species : chr "setosa" "setosa" "setosa" "setosa" ...
View(iris)
# Read in necessary package
# Note that you have to install them first
library(dplyr)
#>
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#>
#> filter, lag
#> The following objects are masked from 'package:base':
#>
#> intersect, setdiff, setequal, union
library(ggplot2)
#> Warning: package 'ggplot2' was built under R version 4.5.3
library(tidyr)1 Univariate
1.1 Categorical and binary variables
Frequency tables
- Count the number of observations per eye-color (frequency table)
- Try to expand this table with percentages and cumulative values
# Base R solution
number <- table(haireye$eye) # number of observations
proportion <- prop.table(table(haireye$eye)) # proportion of observations
cumulative_number <- cumsum(table(haireye$eye)) # cumulative number of observations
cumulative_proportion <- cumsum(prop.table(table(haireye$eye))) # cumulative proportion
table_eye <- rbind(number, proportion, cumulative_number,cumulative_proportion) #combine all over rows
table_eye
#> Blue Brown Green Hazel
#> number 215.0000000 220.0000000 64.0000000 93.0000000
#> proportion 0.3631757 0.3716216 0.1081081 0.1570946
#> cumulative_number 215.0000000 435.0000000 499.0000000 592.0000000
#> cumulative_proportion 0.3631757 0.7347973 0.8429054 1.0000000
# tidyr R solution
freqtab <- haireye %>%
group_by(eye) %>%
summarize(Frequency=n()) %>%
mutate(percent=100*Frequency/sum(Frequency),
cumfreq=cumsum(Frequency),
cumper=cumsum(percent))
freqtab
#> # A tibble: 4 × 5
#> eye Frequency percent cumfreq cumper
#> <chr> <int> <dbl> <int> <dbl>
#> 1 Blue 215 36.3 215 36.3
#> 2 Brown 220 37.2 435 73.5
#> 3 Green 64 10.8 499 84.3
#> 4 Hazel 93 15.7 592 100Bar plot
Try to replicate the barplot.
barplot(table(haireye$hair), main="Barplot for Hair colour", xlab="Hair colour",ylab="Frequency")
# ggplot alternative
haireye %>% ggplot(aes(hair)) + geom_bar() + ggtitle("Barplot of Hair color") + xlab("hair color") + ylab("Frequency")
Try to plot the same barplot with fractions instead of counts at the y-axis
barplot(prop.table(table(haireye$hair)), main="Barplot for Hair colour", xlab="Hair colour",ylab="Frequency")
Try to change the plot so it looks like the example
barplot(table(haireye$hair), main="Barplot for Hair colour", xlab="hair colour",ylab="Frequency",col=c("black","yellow","brown","red"),las=2)
1.2 Continuous variables
Descriptive summary numbers
Produce quartiles, min/max, median, mean, std deviation and variance for the variable Petal.Width in the iris dataset
# Base R solution
quantile(iris$Petal.Width, probs=c(0.25,0.5,0.75)) #Q1, Median, Q3
#> 25% 50% 75%
#> 0.3 1.3 1.8
min(iris$Petal.Width) # Minimum
#> [1] 0.1
max(iris$Petal.Width) # Maximum
#> [1] 2.5
mean(iris$Petal.Width) # Mean
#> [1] 1.199333
sd(iris$Petal.Width) # Standard deviation
#> [1] 0.7622377
var(iris$Petal.Width) # Variance
#> [1] 0.5810063
# Tidyr solution
iris %>% summarise(mean = mean(Petal.Width),
sd = sd(Petal.Width),
min=min(Petal.Width),
max=max(Petal.Width),
lowQu=quantile(Petal.Width,0.25),
median=quantile(Petal.Width,0.5),
upperQu=quantile(Petal.Width,0.75))
#> mean sd min max lowQu median upperQu
#> 1 1.199333 0.7622377 0.1 2.5 0.3 1.3 1.8Box plot
- Try to make a boxplot of Petal Width
- In this boxplot change the title and label as in the example. Plus,
add a text showing the median in the plot
boxplot(iris$Petal.Width, main="Width of Petal leaves", ylab="Leaf width (cm)", col="white")
text(x=median(iris$Petal.Width),pos=2,offset=6,labels="median")
# ggplot alternative
iris %>% ggplot(aes(x=" ", y=Petal.Width)) + geom_boxplot()
Histogram
Produce a histogram showing the frequencies of Iris Sepal Length. Change
the titles and axis-labels to something appropriate
hist(iris$Sepal.Length, main="Length of Iris sepal leaves", xlab="Leaf length (cm)", ylab="Frequency")
# ggplot alternative
iris %>% ggplot(aes(Sepal.Length)) + geom_histogram(binwidth=0.5)+ ggtitle("Histogram of Sepal Length") + xlab("Sepal length") + ylab("Frequency")
Now, try to plot the same histogram with density on the y-axis instead of frequencies. Change the intervals in the x-axis until it looks like the example.
hist(iris$Sepal.Length, freq=F, main="Length of Iris sepal leaves", xlab="Leaf length (cm)", ylab="Density",breaks=20,col="white")
2 Bivariate
2.1 Categorical x categorical variables
Mosaic plot Create a mosaic plot of both variables. Add a title and labels. Also, try to color the tiles according to hair color.
mosaicplot(table(haireye$eye,haireye$hair), main="Mosaic plot", xlab="Eye color", ylab="Hair color", col=c("black","yellow","brown","red"))
# ggplot alternative
df <- haireye %>% group_by(eye,hair) %>% summarise(count = n()) %>% mutate(cut.count = sum(count), prop = count/sum(count))
#> `summarise()` has grouped output by 'eye'. You can override using the `.groups`
#> argument.
ggplot(df,aes(x = eye, y = prop, width = cut.count, fill = hair)) + geom_bar(stat = "identity", position = "fill", colour = "black") + geom_text(aes(label = scales::percent(prop)), position = position_stack(vjust = 0.5)) + facet_grid(~eye, scales = "free_x", space = "free_x") + scale_fill_brewer(palette = "RdYlGn") + theme(panel.spacing.x = unit(0, "npc")) + theme_void()
Stacked barplot - Create a stacked barplot of the
same cross-table with title and labels
- Try to add a legend with the eye-colors so it looks something like the
example.
barplot(table(haireye$eye,haireye$hair),xlab="Hair color",ylab="Frequency", main="Eye color per hair color category",legend.text=c("Blue","Brown","Green","Hazel"))
# ggplot alternative
haireye %>% group_by(eye,hair) %>% summarise(count = n()) %>% ggplot(aes(x = hair, y = count, fill = eye)) + geom_bar(stat ="identity") + xlab("hair color") + ylab("Frequency")
#> `summarise()` has grouped output by 'eye'. You can override using the `.groups`
#> argument.
- Instead of stacked bars, try to group them next to each other and give some colors to the different eye color categories.
barplot(table(haireye$eye,haireye$hair),xlab="Hair color",ylab="Frequency", main="Eye color per hair color category",legend.text=c("Blue","Brown","Green","Hazel"),beside=T,col=c('blue','brown','darkgreen','olivedrab'))
# ggplot alternative
ggplot(haireye, aes(x=hair, fill=eye)) +
geom_bar(position="dodge")
## Categorical x continuous variables
Descriptive summary numbers x categorical levels Generate descriptive summary statistics (min/max, quartiles, mean and median) for Iris Petal Width per level of Iris Species
#Base r solution
tapply(iris$Petal.Width,iris$Species,mean)
#> setosa versicolor virginica
#> 0.246 1.326 2.026
tapply(iris$Petal.Width,iris$Species,min)
#> setosa versicolor virginica
#> 0.1 1.0 1.4
tapply(iris$Petal.Width,iris$Species,max)
#> setosa versicolor virginica
#> 0.6 1.8 2.5
tapply(iris$Petal.Width,iris$Species,quantile,probs=c(0.25,0.5,0.75))
#> $setosa
#> 25% 50% 75%
#> 0.2 0.2 0.3
#>
#> $versicolor
#> 25% 50% 75%
#> 1.2 1.3 1.5
#>
#> $virginica
#> 25% 50% 75%
#> 1.8 2.0 2.3
tapply(iris$Petal.Width,iris$Species,median)
#> setosa versicolor virginica
#> 0.2 1.3 2.0
# Tidyr solution
iris %>%
group_by(Species) %>%
summarise(mean = mean(Petal.Width),
min=min(Petal.Width),
max=max(Petal.Width),
lowQu=quantile(Petal.Width,0.25),
median=quantile(Petal.Width,0.5),
upperQu=quantile(Petal.Width,0.75))
#> # A tibble: 3 × 7
#> Species mean min max lowQu median upperQu
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 setosa 0.246 0.1 0.6 0.2 0.2 0.3
#> 2 versicolor 1.33 1 1.8 1.2 1.3 1.5
#> 3 virginica 2.03 1.4 2.5 1.8 2 2.3Box plots Produce a series of boxplots for the Petal Width by Iris Species in one figure and add title and labels.
boxplot(iris$Sepal.Length~iris$Species, main="Width of Iris Petal leaves", ylab="Leaf length (cm)", col="white", xlab=" ")
# ggplot alternative
iris %>% ggplot(aes(x=Species, y=Sepal.Length)) + geom_boxplot()
2.2 Continuous x continuous variables
Descriptive summary numbers Produce the summary numbers (min/max, quartiles, mean, median) for Sepal.Length, Sepal.Width, Petal.Length, Petal.Width in one go.
summary(iris[,1:4])
#> Sepal.Length Sepal.Width Petal.Length Petal.Width
#> Min. :4.300 Min. :2.000 Min. :1.000 Min. :0.100
#> 1st Qu.:5.100 1st Qu.:2.800 1st Qu.:1.600 1st Qu.:0.300
#> Median :5.800 Median :3.000 Median :4.350 Median :1.300
#> Mean :5.843 Mean :3.057 Mean :3.758 Mean :1.199
#> 3rd Qu.:6.400 3rd Qu.:3.300 3rd Qu.:5.100 3rd Qu.:1.800
#> Max. :7.900 Max. :4.400 Max. :6.900 Max. :2.500Correlation Produce the correlation matrix of these four variables, and try to round all numbers on two decimals
round(cor(iris[,1:4]),2)
#> Sepal.Length Sepal.Width Petal.Length Petal.Width
#> Sepal.Length 1.00 -0.12 0.87 0.82
#> Sepal.Width -0.12 1.00 -0.43 -0.37
#> Petal.Length 0.87 -0.43 1.00 0.96
#> Petal.Width 0.82 -0.37 0.96 1.00Scatter plot
Make a scatterplot for Sepal.Length versus Petal.Length. Try to color
the dots according to species.
plot(iris$Petal.Length,iris$Sepal.Length,col=as.factor(iris$Species), xlab="Petal length (cm)", ylab="Sepal length (cm)")
# ggplot alternative
iris %>% ggplot(aes(x=Petal.Length, y=Sepal.Length, color=Species)) + geom_point()
Now, also try to change the plotting symbol according to Species and add a legend.
plot(iris$Petal.Length,iris$Sepal.Length,col=as.factor(iris$Species), xlab="Petal length (cm)", ylab="Sepal length (cm)", pch=as.numeric(as.factor(iris$Species)))
Scatter plot matrix Create a scatterplot matrix of the same four variables, and color the dots according to species
# BASE R
pairs(iris[,1:4], col=as.factor(iris$Species))