Course Outline

segmentGetting Started (Don't Skip This Part)

segmentIntroduction to Statistics: A Modeling Approach

segmentPART I: EXPLORING VARIATION

segmentChapter 1  Welcome to Statistics: A Modeling Approach

segmentChapter 2  Understanding Data

segmentChapter 3  Examining Distributions

3.7 Boxplots and the FiveNumber Summary

segmentChapter 4  Explaining Variation

segmentPART II: MODELING VARIATION

segmentChapter 5  A Simple Model

segmentChapter 6  Quantifying Error

segmentChapter 7  Adding an Explanatory Variable to the Model

segmentChapter 8  Models with a Quantitative Explanatory Variable

segmentPART III: EVALUATING MODELS

segmentChapter 9  Distributions of Estimates

segmentChapter 10  Confidence Intervals and Their Uses

segmentChapter 11  Model Comparison with the F Ratio

segmentChapter 12  What You Have Learned

segmentResources
list Introduction to Statistics: A Modeling Approach
Boxplots and the FiveNumber Summary
Boxplots are a handy tool for visualizing the fivenumber summary of a distribution. Making boxplots with the function gf_boxplot()
will also clearly show you the IQR and outliers! Very handy.
Unlike histograms, where the values of the variable went on the xaxis, the boxplots made with gf_boxplot()
put the values of the variable on the yaxis. Boxplots do not have to be made this way; this is just the way it is done by gf_boxplot()
.
L_Ch3_Boxplots_9
Here is the code for making a boxplot of Wt from MindsetMatters with gf_boxplot()
.
gf_boxplot(Wt ~ 1, data = MindsetMatters)
The 1 just means that there is only going to be one boxplot here. Later we will replace that as we explore methods of making multiple boxplots that appear next to each other.
The boxplot is made up of a few parts. There is a big white box with two parts–an upper and lower part. There are lines, called whiskers, above and below the box. Another name for boxplot is boxandwhisker plot.
This is a case where there are no outliers (defined as more than 1.5 IQRs above Q3 or below Q1). So the whiskers will simply end at the max and min values for Wt.
L_Ch3_Boxplots_10
Modify this code to create a boxplot for Population from the HappyPlanetIndex data frame.
require(mosaic)
require(tidyverse)
require(Lock5withR)
require(supernova)
HappyPlanetIndex$Region < recode(HappyPlanetIndex$Region, '1'="Latin America", '2'="Western Nations", '3'="Middle East and North Africa", '4'="SubSaharan Africa", '5'="South Asia", '6'="East Asia", '7'="Former Communist Countries")
# Modify this code to create a boxplot of Population from HappyPlanetIndex
gf_boxplot(Wt ~ 1, data=MindsetMatters)
# Modify this code to create a boxplot of Population from HappyPlanetIndex
gf_boxplot(Population ~ 1, data=HappyPlanetIndex)
test_function("gf_boxplot", args="data", incorrect_msg="Did you change the dataset to `HappyPlanetIndex`?")
test_function_result("gf_boxplot", incorrect_msg="Did you create a boxplot of `Population`?")
test_error()
success_msg("Great work!")
Wow, this is a strange looking boxplot. You can hardly see the box; it’s squished down on the bottom. And there are all these points here even though it’s supposed to be depicting a box and whisker plot.
The points that appear on a boxplot are the outliers. If they appear above the top whisker, they are outliers because R has checked whether these values are greater than the \(Q3 + 1.5*IQR\). If they appear below the bottom whisker, they are outliers because their values are smaller than the \(Q1  1.5*IQR\). When there are outliers, the end of the whisker depicts the max or min value that is not considered an outlier.
L_Ch3_Boxplots_11
There are a lot of large outlier countries. No wonder the histogram we looked at before put so many countries into the same bin! It looks as though most countries are at 0 millions. If only we could “zoom in” on these countries with a smaller population.
L_Ch3_Boxplots_12
In the following DataCamp window, use filter()
to get just the countries with populations smaller than this upper boundary. Save these countries in a data frame called SmallerCountries. Run the code to see a histogram of those Population data.
require(mosaic)
require(tidyverse)
require(Lock5withR)
require(supernova)
HappyPlanetIndex$Region < recode(HappyPlanetIndex$Region, '1'="Latin America", '2'="Western Nations", '3'="Middle East and North Africa", '4'="SubSaharan Africa", '5'="South Asia", '6'="East Asia", '7'="Former Communist Countries")
# this calculates the Q3 + 1.5*IQR
UpperBoundary < 31.225 + 1.5*(31.2254.455)
# modify this code to filter in only countries with population sizes less than the UpperBoundary
SmallerCountries <
# this makes a histogram of the smaller countries' populations
gf_histogram(~ Population, data = SmallerCountries, fill = "slateblue4") %>%
gf_labs(x = "Population (in millions)", title = "Population of Countries (Excludes Outliers)")
# this calculates the Q3 + 1.5*IQR
UpperBoundary < 31.225 + 1.5*(31.2254.455)
# modify this code for Population from HappyPlanetIndex
SmallerCountries < filter(HappyPlanetIndex, Population < UpperBoundary)
# this makes a histogram of the smaller countries' populations
gf_histogram(~ Population, data = SmallerCountries, fill = "slateblue4") %>%
gf_labs(x = "Population (in millions)", title = "Population of Countries (Excludes Outliers)")
ex() %>% check_object("UpperBoundary") %>% check_equal("UpperBoundary")
ex() %>% check_function("filter") %>% check_result() %>% check_equal()
ex() %>% check_function("filter") %>% check_arg(".data") %>% check_equal(incorrect_msg="Don't forget to filter in HappyPlanetIndex")
ex() %>% check_function("filter") %>% check_arg("...") %>% check_equal(incorrect_msg="Did you use `Population < UpperBoundary` as the second argument?")
ex() %>% check_object("SmallerCountries") %>% check_equal()
ex() %>% check_function("gf_histogram") %>% check_arg("object") %>% check_equal()
ex() %>% check_function("gf_histogram") %>% check_arg("data") %>% check_equal()
ex() %>% check_function("gf_labs") %>% check_arg("x") %>% check_equal()
ex() %>% check_function("gf_labs") %>% check_arg("title") %>% check_equal()
ex() %>% check_error()
success_msg("That was a tough one! Great job working through it.")
Ah, this is a very different histogram than the one that included outliers. Here we get a sense of how the countries that previously got lumped together in one bin actually vary in their population size.
L_Ch3_Boxplots_13
Let’s rerun the boxplot for just these countries in the data frame SmallerCountries to see what that looks like. Just press the Run button.
require(mosaic)
require(tidyverse)
require(Lock5withR)
require(supernova)
HappyPlanetIndex$Region < recode(HappyPlanetIndex$Region, '1'="Latin America", '2'="Western Nations", '3'="Middle East and North Africa", '4'="SubSaharan Africa", '5'="South Asia", '6'="East Asia", '7'="Former Communist Countries")
Pop.stats < favstats(~ Population, data = HappyPlanetIndex)
SmallerCountries < filter(HappyPlanetIndex, Population < (Pop.stats$Q3 + 1.5*(Pop.stats$Q3  Pop.stats$Q1)))
# Make a boxplot of Population from the SmallerCountries
gf_boxplot(Population ~ 1, data=SmallerCountries)
# Make a boxplot of Population from the SmallerCountries
gf_boxplot(Population ~ 1, data=SmallerCountries)
test_function_result("gf_boxplot")
test_error()
success_msg("Nice work!")
L_Ch3_Boxplots_14