In graebnerc/DataScienceExercises: Exercises for introductions to R
library(learnr)
library(gradethis)
library(ggcheck)
library(ggplot2)
library(scales)
library(gradethis)
knitr::opts_chunk$set(echo = FALSE)
gradethis_setup(
pass.praise = TRUE, fail.hint = FALSE, fail_code_feedback = FALSE,
fail.encourage = TRUE, maybe_code_feedback = FALSE)
nycflights21_small <- DataScienceExercises::nycflights21_small
nycflights21_small <- dplyr::mutate(nycflights21_small, month=factor(month))
set.seed(123)
rdn_indices <- sample(1:nrow(nycflights21_small), 10000, replace = FALSE)
nycflights21_small <- nycflights21_small[rdn_indices, ]
Content quiz
quiz(
question(
paste0(
"When choosing the right graph you should think about what your graph ",
"should achieve. Which are the four main goals discussed in the ",
"lecture?"
),
answer('Comparison', correct = TRUE),
answer('Distribution', correct = TRUE),
answer('Composition', correct = TRUE),
answer('Relationship', correct = TRUE),
answer('Accuracy'),
answer('Consistency'),
answer('Measurement'),
allow_retry = TRUE, random_answer_order = TRUE
),
question(
paste0(
"Suppose you defined an object using `ggplot2::ggplot()`. ",
"What do you need to do to actually render the plot?"
),
answer('Call the object or save it via `ggplot2::ggsave()`', correct = TRUE),
answer('Nothing, `ggplot2::ggplot()` already does the rendering for you'),
answer('Call `ggplot2::gg_render()` with the plot as the main argument'),
answer('Use R-Studios `Save as` functionality'),
allow_retry = TRUE, random_answer_order = TRUE
),
question(
paste0(
"Aesthetics are visual properties of plots that are mapped to variables ",
"via the function `ggplot2::aes()` or set manually.",
"Which of the following things are aesthetics?"
),
answer('Size', correct = TRUE),
answer('Shape', correct = TRUE),
answer('Color', correct = TRUE),
answer('Fill', correct = TRUE),
answer('Point'),
answer('Bar'),
answer('Line'),
answer('Smooth'),
allow_retry = TRUE, random_answer_order = TRUE
),
question(
paste0(
"Geometrical objects used to represent data are called 'geoms' in the ",
"`ggplot2` context and are added to a plot via `ggplot2::geom_*()`.",
"What of the following things can be used as geoms?"
),
answer('Size'),
answer('Shape'),
answer('Color'),
answer('Fill'),
answer('Point', correct = TRUE),
answer('Bar', correct = TRUE),
answer('Line', correct = TRUE),
answer('Smooth', correct = TRUE),
allow_retry = TRUE, random_answer_order = TRUE
),
question(
"What are 'facets' used for?",
answer(paste0(
'They can complement or substitute aesthetic mappings:",
"you can split your plot into facets, i.e. subplots that each",
"display one subset of the data.'),
correct = TRUE),
answer('Point'),
answer('Bar'),
answer('Line'),
answer('Smooth'),
allow_retry = TRUE, random_answer_order = TRUE
),
question(
"What is true about the relationship between geoms and aesthetics?",
answer(paste0(
'Not every aesthetic works with every geom and vice versa.'),
correct = TRUE),
answer('Every aesthetic can be combined, in principle, with every geom.'),
answer(paste0(
"Each aesthetic can be used to specify any geom, but not vice versa.")),
answer(paste0(
'Aesthetics and geoms and two substitutive ways to map data onto ',
'visual properties of a plot.')),
answer(paste0(
'Since aesthetics and geoms are used for different purposes, there is no ',
'meaningful relationship between them.')),
allow_retry = TRUE, random_answer_order = TRUE
)
)
Coding exercise - Part 1
Note:
Automated feedback does not yet work well with visualization tasks.
Thus, for some exercises you must compare your result with the solution
yourself. These are the exercises that are lacking the Submit Answer button.
The following data set called nycflights21_small contains information about
flights from the New York City airports in 2021.
head(nycflights21_small)
The columns mean the following:
dep_delay, arr_delay: Departure and arrival delays, in minutes. Negative times * represent early departures/arrivals.month: Month of departure.carrier: Two letter carrier abbreviation.distance: Distance between airports, in miles.
Start by producing a very simple plot where you simply take the data set, and
represent the relationship between arr_delay and dep_delay using a
scatterplot. Here, arr_delay should go on the y-axis, dep_delay on
the x-axis.
ggplot(
data = ____,
mapping = ____
) +
geom_____()
ggplot(
data = ____,
mapping = aes(____)
) +
geom_____()
ggplot(
data = nycflights21_small,
mapping = aes(x=____, y=____)
) +
geom____()
ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y_____)
) +
geom_point()
ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay)
) +
geom_point()
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE),
message = paste0(
"To produce a scatterplot you must use the geom 'point'."
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=dep_delay, y=arr_delay), exact = TRUE),
message = paste0(
"You should map the variable `dep_delay` to the x-axis, and the ",
"variable `arr_delay` to the y-axis! Use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
pass()
}
)
Create the same plot as in the exercise before, but (1) map the variable carrier
to the aesthetic 'color', and (2) make sure the points have a transparency of 50%.
ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, ____)
) +
geom_point(____)
ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, color=____)
) +
geom_point(____)
ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, color=____)
) +
geom_point(alpha=____)
ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, color=____)
) +
geom_point(alpha=0.5)
ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, color=carrier)
) +
geom_point(alpha=0.5)
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE),
message = paste0(
"To produce a scatterplot you must use the geom 'point'."
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=dep_delay, y=arr_delay, color=carrier),
exact = TRUE),
message = paste0(
"You should map the variable `dep_delay` to the x-axis, the ",
"variable `arr_delay` to the y-axis, and the variable `carrier` to ",
"the color! Use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
fail_if(!uses_geom_params(.result, "point", list(alpha = 0.5)),
message = paste0(
"Your point are not fifty percent transparent. You can set ",
"transparency within the geom call. Google the relevant aesthetic!"
)
)
pass()
})
Assume the plot from the previous exercise was saved with the name plot_1.
Adjust the following labels:
- Title should be: "Delays in departure and arrival in 2021"
- Title of x-axis should be: "Departure delay (in min.)"
- Title of y-axis should be: "Arrival delay (in min.)"
- Caption should be: "Data: anyflights package."
plot_1 <- ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, color=carrier)
) +
geom_point(alpha=0.5)
plot_1 +
____(
____ = ____,
____ = ____,
____ = ____,
____ = ____
)
plot_1 +
labs(
____ = ____,
____ = ____,
____ = ____,
____ = ____
)
plot_1 +
labs(
title = ____,
____ = ____,
____ = ____,
____ = ____
)
plot_1 +
____(
title = "Delays in departure and arrival in 2021",
____ = ____,
____ = ____,
____ = ____
)
plot_1 +
____(
title = "Delays in departure and arrival in 2021",
x = ____,
y = ____,
caption = ____
)
plot_1 +
labs(
title = "Delays in departure and arrival in 2021",
x = "Departure delay (in min.)",
y = "Arrival delay (in min.)",
caption = "Data: anyflights package."
)
x_label <- "Departure delay (in min.)"
y_label <- "Arrival delay (in min.)"
title_label <- "Delays in departure and arrival in 2021"
cap_label <- "Data: anyflights package."
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`.",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE),
message = paste0(
"To produce a scatterplot you must use the geom 'point'."
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=dep_delay, y=arr_delay, color=carrier),
exact = TRUE),
message = paste0(
"You should map the variable `dep_delay` to the x-axis, the ",
"variable `arr_delay` to the y-axis, and the variable `carrier` to ",
"the color! Use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
res_labs <- get_labels(.result)
fail_if(is.null(res_labs$x),
message = paste0(
"Label for x axis not set. Try using the `ggplot2::labs()` function."))
fail_if(is.null(res_labs$y),
message = paste0(
"Label for y axis not set. Try using the `ggplot2::labs()` function."))
fail_if(res_labs$x != x_label,
message = paste0(
"The label of the x-axis should be '", x_label, "', ",
"not '{res_labs$x}' as in your case!"))
fail_if(res_labs$y != y_label,
message = paste0(
"The label of the y-axis should be '", y_label, "', ",
"not '{res_labs$y}' as in your case!"))
fail_if(is.null(res_labs$caption),
message = paste0(
"Caption not set. Try using the `ggplot2::labs()` function."))
fail_if(is.null(res_labs$title),
message = paste0(
"Title not set. Try using the `ggplot2::labs()` function."))
fail_if(res_labs$title != title_label,
message = paste0(
"The caption should be '", title_label, "', ",
"not '{res_labs$title}' as in your case!"))
fail_if(res_labs$caption != cap_label,
message = paste0(
"The caption should be '", cap_label, "', ",
"not '{res_labs$caption}' as in your case!"))
fail_if(res_labs$title != title_label,
message = paste0(
"The caption should be '", title_label, "', ",
"not '{res_labs$title}' as in your case!"))
pass()
}
)
Again, assume the plot from the previous exercise is saved as plot_1.
Now adjust the x- and y-axis labels: they should be divided by 1000, and a
suffix 'k' should be added.
Note: No automatic feedback available for this exercise, you need to
compare your solution to the provided solution yourself by clicking through
all hints.
plot_1 <- ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, color=carrier)
) +
geom_point(alpha=0.5) +
labs(
title = "Delays in departure and arrival in 2021",
x = "Departure delay (in min.)",
y = "Arrival delay (in min.)",
caption = "Data: anyflights package."
)
plot_1 +
scale_____(
____) +
scale_____(
____)
plot_1 +
scale_x____(
____) +
scale_y____(
____)
plot_1 +
scale_x_continuous(
____) +
scale_y_continuous(
____)
plot_1 +
scale_x_continuous(
labels = ____(____ = ____, ____ = ____)) +
scale_y_continuous(
____)
plot_1 +
scale_x_continuous(
labels = scales::number_format(____ = ____, ____ = ____)) +
scale_y_continuous(
____)
plot_1 +
scale_x_continuous(
labels = scales::number_format(scale = ____, ____ = ____)) +
scale_y_continuous(
____)
plot_1 +
scale_y_continuous(
labels = scales::number_format(scale = ____, suffix = ____)) +
scale_x_continuous(
____)
plot_1 +
scale_y_continuous(
labels = scales::number_format(scale = 0.001, suffix = "k")) +
scale_x_continuous(
____)
plot_1 +
scale_y_continuous(
labels = scales::number_format(scale = 0.001, suffix = "k")) +
scale_x_continuous(
labels = scales::number_format(scale = 0.001, suffix = "k"))
Finally, apply the black-and-white theme to the plot plot_1.
Note: No automatic feedback available for this exercise, you need to
compare your solution to the provided solution yourself by clicking through
all hints.
plot_1 <- ggplot(
data = nycflights21_small,
mapping = aes(x=dep_delay, y = arr_delay, color=carrier)
) +
geom_point(alpha=0.5) +
labs(
title = "Delays in departure and arrival in 2021",
x = "Departure delay (in min.)",
y = "Arrival delay (in min.)",
caption = "Data: anyflights package."
) +
scale_y_continuous(
labels = scales::number_format(scale = 0.001, suffix = "k")) +
scale_x_continuous(
labels = scales::number_format(scale = 0.001, suffix = "k"))
plot_1 +
____()
plot_1 +
theme_____()
plot_1 +
theme_bw()
Coding exercise - Part 2
Let us now study how arrival delays are distributed over the year.
To this end, produce a plot by mapping the variable month to the x-axis, and
the variable arr_delay to the y axis. The data should be represented via the
point-geom. Also apply the black-and-white theme.
ggplot(
data = ____,
mapping = aes(____)
) +
____() +
____()
ggplot(
data = ____,
mapping = aes(____)
) +
____() +
theme_bw()
ggplot(
data = ____,
mapping = aes(x=____, y = ____)
) +
____() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=____, y = ____)
) +
____() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=____, y = ____)
) +
geom_() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=____, y = ____)
) +
geom_point() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = ____)
) +
geom_point() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_point() +
theme_bw()
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE),
message = paste0(
"You were asked to use the geom 'point'."
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay),
exact = TRUE),
message = paste0(
"You should map the variable `month` to the x-axis, and the ",
"variable `arr_delay` to the y-axis!" ,
"Use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
pass()
})
One way to represent distributions with more information is to use a
boxplot.
You can do this by replacing the geom 'point' with the geom 'boxplot'.
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
____() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_____() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_boxplot() +
theme_bw()
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_geoms(.result, "boxplot", exact = FALSE),
message = paste0(
"You were asked to use the geom 'boxplot'."
), encourage = TRUE
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay),
exact = TRUE),
message = paste0(
"You should map the variable `month` to the x-axis, and the ",
"variable `arr_delay` to the y-axis!" ,
"Use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
pass()
})
One problem with boxplots is that they may conceal the underlying distribution
of the data. A neat alternative is the
violin plot.
Replace the boxplot with the geom needed for a violin plot.
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_____() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_violin() +
theme_bw()
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_geoms(.result, "violin", exact = FALSE),
message = paste0(
"You were asked to use the geom 'violin'."
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay),
exact = TRUE),
message = paste0(
"You should map the variable `month` to the x-axis, and the ",
"variable `arr_delay` to the y-axis!" ,
"Use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
pass()
})
Redo the previous plot but set the optional argument for the violin geom
scale to 'count'. Can you guess what it does?
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_violin(____) +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_violin(scale = ____) +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x=month, y = arr_delay)
) +
geom_violin(scale = "count") +
theme_bw()
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay),
exact = TRUE),
message = paste0(
"You should map the variable `month` to the x-axis, and the ",
"variable `arr_delay` to the y-axis!" ,
"Use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
fail_if(!ggcheck::uses_geoms(.result, "violin", exact = FALSE),
message = paste0(
"You were asked to use the geom 'violin'."
)
)
fail_if(!uses_geom_params(.result, "violin", list(scale = "count")),
message = paste0(
"You forgot the set the optional argument `scale` for the violin ",
"geom to `'count`. Google can help you to find out what to do!"
)
)
pass(message = paste0("Great job! As you can see the specification ",
"`scale = 'count'` is useful to highlight differences ",
"in the sizes of the different groups that otherwise ",
"might remain unnoticed!"))
})
Finally, lets look at the distribution of delays. We might do this using a
histogram. Create a histogram of the arrival delays.
Hint: you now need to specify a single axis, since you are representing
only the distribution of a single variable.
ggplot(
data = nycflights21_small,
mapping = aes(____)
) +
____() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x = ____)
) +
____() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x = ____)
) +
geom_histogram() +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x = arr_delay)
) +
geom_histogram() +
theme_bw()
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=arr_delay),
exact = TRUE),
message = paste0(
"There is a problem with your mapping. You only need one variable " ,
"this time. Think about which axis it would make sense to plot it ",
"onto. As usual, use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
fail_if(!ggcheck::uses_geoms(.result, "histogram", exact = FALSE),
message = paste0(
"You must produce a histogram. What geom could be the right one?"
)
)
pass()
})
Re-create this histogram from above, but set the width of the histogram bins
to 10.
ggplot(
data = nycflights21_small,
mapping = aes(x = arr_delay)
) +
geom_histogram(____) +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x = arr_delay)
) +
geom_histogram(____ = 10) +
theme_bw()
ggplot(
data = nycflights21_small,
mapping = aes(x = arr_delay)
) +
geom_histogram(binwidth = 10) +
theme_bw()
grade_this({
fail_if(!ggcheck::is_ggplot(.result),
message = paste0(
"You did not define a `ggplot` object. Use the function ",
"`ggplot2::ggplot()`. ",
"Did you maybe forget to call and render the plot?"
)
)
fail_if(!ggcheck::uses_data(.result, nycflights21_small),
message = paste0(
"You should use the data set `nycflights21_small`. You can pass ",
"it to the argument `data` of `ggplot2::ggplot()`."
)
)
fail_if(!uses_mappings(.result, aes(x=arr_delay),
exact = TRUE),
message = paste0(
"There is a problem with your mapping. You only need one variable " ,
"this time. Think about which axis it would make sense to plot it ",
"onto. As usual, use the argument `mapping` ",
"of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`."
)
)
fail_if(!ggcheck::uses_geoms(.result, "histogram", exact = FALSE),
message = paste0(
"You must produce a histogram. What geom could be the right one?"
)
)
fail_if(!uses_geom_params(.result, "histogram", list(binwidth = 10)),
message = paste0(
"You forgot the set the binwidth of the histogram. It might be ",
"a good idea to google how to set the binwidth (not the nb of bins!)!"
)
)
pass()
})
graebnerc/DataScienceExercises documentation built on April 11, 2025, 7:57 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
library(learnr) library(gradethis) library(ggcheck) library(ggplot2) library(scales) library(gradethis) knitr::opts_chunk$set(echo = FALSE) gradethis_setup( pass.praise = TRUE, fail.hint = FALSE, fail_code_feedback = FALSE, fail.encourage = TRUE, maybe_code_feedback = FALSE) nycflights21_small <- DataScienceExercises::nycflights21_small nycflights21_small <- dplyr::mutate(nycflights21_small, month=factor(month)) set.seed(123) rdn_indices <- sample(1:nrow(nycflights21_small), 10000, replace = FALSE) nycflights21_small <- nycflights21_small[rdn_indices, ]
Content quiz
quiz( question( paste0( "When choosing the right graph you should think about what your graph ", "should achieve. Which are the four main goals discussed in the ", "lecture?" ), answer('Comparison', correct = TRUE), answer('Distribution', correct = TRUE), answer('Composition', correct = TRUE), answer('Relationship', correct = TRUE), answer('Accuracy'), answer('Consistency'), answer('Measurement'), allow_retry = TRUE, random_answer_order = TRUE ), question( paste0( "Suppose you defined an object using `ggplot2::ggplot()`. ", "What do you need to do to actually render the plot?" ), answer('Call the object or save it via `ggplot2::ggsave()`', correct = TRUE), answer('Nothing, `ggplot2::ggplot()` already does the rendering for you'), answer('Call `ggplot2::gg_render()` with the plot as the main argument'), answer('Use R-Studios `Save as` functionality'), allow_retry = TRUE, random_answer_order = TRUE ), question( paste0( "Aesthetics are visual properties of plots that are mapped to variables ", "via the function `ggplot2::aes()` or set manually.", "Which of the following things are aesthetics?" ), answer('Size', correct = TRUE), answer('Shape', correct = TRUE), answer('Color', correct = TRUE), answer('Fill', correct = TRUE), answer('Point'), answer('Bar'), answer('Line'), answer('Smooth'), allow_retry = TRUE, random_answer_order = TRUE ), question( paste0( "Geometrical objects used to represent data are called 'geoms' in the ", "`ggplot2` context and are added to a plot via `ggplot2::geom_*()`.", "What of the following things can be used as geoms?" ), answer('Size'), answer('Shape'), answer('Color'), answer('Fill'), answer('Point', correct = TRUE), answer('Bar', correct = TRUE), answer('Line', correct = TRUE), answer('Smooth', correct = TRUE), allow_retry = TRUE, random_answer_order = TRUE ), question( "What are 'facets' used for?", answer(paste0( 'They can complement or substitute aesthetic mappings:", "you can split your plot into facets, i.e. subplots that each", "display one subset of the data.'), correct = TRUE), answer('Point'), answer('Bar'), answer('Line'), answer('Smooth'), allow_retry = TRUE, random_answer_order = TRUE ), question( "What is true about the relationship between geoms and aesthetics?", answer(paste0( 'Not every aesthetic works with every geom and vice versa.'), correct = TRUE), answer('Every aesthetic can be combined, in principle, with every geom.'), answer(paste0( "Each aesthetic can be used to specify any geom, but not vice versa.")), answer(paste0( 'Aesthetics and geoms and two substitutive ways to map data onto ', 'visual properties of a plot.')), answer(paste0( 'Since aesthetics and geoms are used for different purposes, there is no ', 'meaningful relationship between them.')), allow_retry = TRUE, random_answer_order = TRUE ) )
Coding exercise - Part 1
Note:
Automated feedback does not yet work well with visualization tasks.
Thus, for some exercises you must compare your result with the solution
yourself. These are the exercises that are lacking the Submit Answer button.
The following data set called nycflights21_small contains information about
flights from the New York City airports in 2021.
head(nycflights21_small)
The columns mean the following:
dep_delay,arr_delay: Departure and arrival delays, in minutes. Negative times * represent early departures/arrivals.month: Month of departure.carrier: Two letter carrier abbreviation.distance: Distance between airports, in miles.
Start by producing a very simple plot where you simply take the data set, and
represent the relationship between arr_delay and dep_delay using a
scatterplot. Here, arr_delay should go on the y-axis, dep_delay on
the x-axis.
ggplot( data = ____, mapping = ____ ) + geom_____()
ggplot( data = ____, mapping = aes(____) ) + geom_____()
ggplot( data = nycflights21_small, mapping = aes(x=____, y=____) ) + geom____()
ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y_____) ) + geom_point()
ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay) ) + geom_point()
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE), message = paste0( "To produce a scatterplot you must use the geom 'point'." ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=dep_delay, y=arr_delay), exact = TRUE), message = paste0( "You should map the variable `dep_delay` to the x-axis, and the ", "variable `arr_delay` to the y-axis! Use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) pass() } )
Create the same plot as in the exercise before, but (1) map the variable carrier
to the aesthetic 'color', and (2) make sure the points have a transparency of 50%.
ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, ____) ) + geom_point(____)
ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, color=____) ) + geom_point(____)
ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, color=____) ) + geom_point(alpha=____)
ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, color=____) ) + geom_point(alpha=0.5)
ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, color=carrier) ) + geom_point(alpha=0.5)
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE), message = paste0( "To produce a scatterplot you must use the geom 'point'." ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=dep_delay, y=arr_delay, color=carrier), exact = TRUE), message = paste0( "You should map the variable `dep_delay` to the x-axis, the ", "variable `arr_delay` to the y-axis, and the variable `carrier` to ", "the color! Use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) fail_if(!uses_geom_params(.result, "point", list(alpha = 0.5)), message = paste0( "Your point are not fifty percent transparent. You can set ", "transparency within the geom call. Google the relevant aesthetic!" ) ) pass() })
Assume the plot from the previous exercise was saved with the name plot_1.
Adjust the following labels:
- Title should be: "Delays in departure and arrival in 2021"
- Title of x-axis should be: "Departure delay (in min.)"
- Title of y-axis should be: "Arrival delay (in min.)"
- Caption should be: "Data: anyflights package."
plot_1 <- ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, color=carrier) ) + geom_point(alpha=0.5)
plot_1 + ____( ____ = ____, ____ = ____, ____ = ____, ____ = ____ )
plot_1 + labs( ____ = ____, ____ = ____, ____ = ____, ____ = ____ )
plot_1 + labs( title = ____, ____ = ____, ____ = ____, ____ = ____ )
plot_1 + ____( title = "Delays in departure and arrival in 2021", ____ = ____, ____ = ____, ____ = ____ )
plot_1 + ____( title = "Delays in departure and arrival in 2021", x = ____, y = ____, caption = ____ )
plot_1 + labs( title = "Delays in departure and arrival in 2021", x = "Departure delay (in min.)", y = "Arrival delay (in min.)", caption = "Data: anyflights package." )
x_label <- "Departure delay (in min.)" y_label <- "Arrival delay (in min.)" title_label <- "Delays in departure and arrival in 2021" cap_label <- "Data: anyflights package." grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`.", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE), message = paste0( "To produce a scatterplot you must use the geom 'point'." ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=dep_delay, y=arr_delay, color=carrier), exact = TRUE), message = paste0( "You should map the variable `dep_delay` to the x-axis, the ", "variable `arr_delay` to the y-axis, and the variable `carrier` to ", "the color! Use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) res_labs <- get_labels(.result) fail_if(is.null(res_labs$x), message = paste0( "Label for x axis not set. Try using the `ggplot2::labs()` function.")) fail_if(is.null(res_labs$y), message = paste0( "Label for y axis not set. Try using the `ggplot2::labs()` function.")) fail_if(res_labs$x != x_label, message = paste0( "The label of the x-axis should be '", x_label, "', ", "not '{res_labs$x}' as in your case!")) fail_if(res_labs$y != y_label, message = paste0( "The label of the y-axis should be '", y_label, "', ", "not '{res_labs$y}' as in your case!")) fail_if(is.null(res_labs$caption), message = paste0( "Caption not set. Try using the `ggplot2::labs()` function.")) fail_if(is.null(res_labs$title), message = paste0( "Title not set. Try using the `ggplot2::labs()` function.")) fail_if(res_labs$title != title_label, message = paste0( "The caption should be '", title_label, "', ", "not '{res_labs$title}' as in your case!")) fail_if(res_labs$caption != cap_label, message = paste0( "The caption should be '", cap_label, "', ", "not '{res_labs$caption}' as in your case!")) fail_if(res_labs$title != title_label, message = paste0( "The caption should be '", title_label, "', ", "not '{res_labs$title}' as in your case!")) pass() } )
Again, assume the plot from the previous exercise is saved as plot_1.
Now adjust the x- and y-axis labels: they should be divided by 1000, and a
suffix 'k' should be added.
Note: No automatic feedback available for this exercise, you need to compare your solution to the provided solution yourself by clicking through all hints.
plot_1 <- ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, color=carrier) ) + geom_point(alpha=0.5) + labs( title = "Delays in departure and arrival in 2021", x = "Departure delay (in min.)", y = "Arrival delay (in min.)", caption = "Data: anyflights package." )
plot_1 + scale_____( ____) + scale_____( ____)
plot_1 + scale_x____( ____) + scale_y____( ____)
plot_1 + scale_x_continuous( ____) + scale_y_continuous( ____)
plot_1 + scale_x_continuous( labels = ____(____ = ____, ____ = ____)) + scale_y_continuous( ____)
plot_1 + scale_x_continuous( labels = scales::number_format(____ = ____, ____ = ____)) + scale_y_continuous( ____)
plot_1 + scale_x_continuous( labels = scales::number_format(scale = ____, ____ = ____)) + scale_y_continuous( ____)
plot_1 + scale_y_continuous( labels = scales::number_format(scale = ____, suffix = ____)) + scale_x_continuous( ____)
plot_1 + scale_y_continuous( labels = scales::number_format(scale = 0.001, suffix = "k")) + scale_x_continuous( ____)
plot_1 + scale_y_continuous( labels = scales::number_format(scale = 0.001, suffix = "k")) + scale_x_continuous( labels = scales::number_format(scale = 0.001, suffix = "k"))
Finally, apply the black-and-white theme to the plot plot_1.
Note: No automatic feedback available for this exercise, you need to compare your solution to the provided solution yourself by clicking through all hints.
plot_1 <- ggplot( data = nycflights21_small, mapping = aes(x=dep_delay, y = arr_delay, color=carrier) ) + geom_point(alpha=0.5) + labs( title = "Delays in departure and arrival in 2021", x = "Departure delay (in min.)", y = "Arrival delay (in min.)", caption = "Data: anyflights package." ) + scale_y_continuous( labels = scales::number_format(scale = 0.001, suffix = "k")) + scale_x_continuous( labels = scales::number_format(scale = 0.001, suffix = "k"))
plot_1 + ____()
plot_1 + theme_____()
plot_1 + theme_bw()
Coding exercise - Part 2
Let us now study how arrival delays are distributed over the year.
To this end, produce a plot by mapping the variable month to the x-axis, and
the variable arr_delay to the y axis. The data should be represented via the
point-geom. Also apply the black-and-white theme.
ggplot( data = ____, mapping = aes(____) ) + ____() + ____()
ggplot( data = ____, mapping = aes(____) ) + ____() + theme_bw()
ggplot( data = ____, mapping = aes(x=____, y = ____) ) + ____() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=____, y = ____) ) + ____() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=____, y = ____) ) + geom_() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=____, y = ____) ) + geom_point() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=month, y = ____) ) + geom_point() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_point() + theme_bw()
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_geoms(.result, "point", exact = FALSE), message = paste0( "You were asked to use the geom 'point'." ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay), exact = TRUE), message = paste0( "You should map the variable `month` to the x-axis, and the ", "variable `arr_delay` to the y-axis!" , "Use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) pass() })
One way to represent distributions with more information is to use a boxplot. You can do this by replacing the geom 'point' with the geom 'boxplot'.
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + ____() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_____() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_boxplot() + theme_bw()
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_geoms(.result, "boxplot", exact = FALSE), message = paste0( "You were asked to use the geom 'boxplot'." ), encourage = TRUE ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay), exact = TRUE), message = paste0( "You should map the variable `month` to the x-axis, and the ", "variable `arr_delay` to the y-axis!" , "Use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) pass() })
One problem with boxplots is that they may conceal the underlying distribution of the data. A neat alternative is the violin plot. Replace the boxplot with the geom needed for a violin plot.
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_____() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_violin() + theme_bw()
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_geoms(.result, "violin", exact = FALSE), message = paste0( "You were asked to use the geom 'violin'." ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay), exact = TRUE), message = paste0( "You should map the variable `month` to the x-axis, and the ", "variable `arr_delay` to the y-axis!" , "Use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) pass() })
Redo the previous plot but set the optional argument for the violin geom
scale to 'count'. Can you guess what it does?
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_violin(____) + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_violin(scale = ____) + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x=month, y = arr_delay) ) + geom_violin(scale = "count") + theme_bw()
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=month, y = arr_delay), exact = TRUE), message = paste0( "You should map the variable `month` to the x-axis, and the ", "variable `arr_delay` to the y-axis!" , "Use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) fail_if(!ggcheck::uses_geoms(.result, "violin", exact = FALSE), message = paste0( "You were asked to use the geom 'violin'." ) ) fail_if(!uses_geom_params(.result, "violin", list(scale = "count")), message = paste0( "You forgot the set the optional argument `scale` for the violin ", "geom to `'count`. Google can help you to find out what to do!" ) ) pass(message = paste0("Great job! As you can see the specification ", "`scale = 'count'` is useful to highlight differences ", "in the sizes of the different groups that otherwise ", "might remain unnoticed!")) })
Finally, lets look at the distribution of delays. We might do this using a histogram. Create a histogram of the arrival delays.
Hint: you now need to specify a single axis, since you are representing only the distribution of a single variable.
ggplot( data = nycflights21_small, mapping = aes(____) ) + ____() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x = ____) ) + ____() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x = ____) ) + geom_histogram() + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x = arr_delay) ) + geom_histogram() + theme_bw()
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=arr_delay), exact = TRUE), message = paste0( "There is a problem with your mapping. You only need one variable " , "this time. Think about which axis it would make sense to plot it ", "onto. As usual, use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) fail_if(!ggcheck::uses_geoms(.result, "histogram", exact = FALSE), message = paste0( "You must produce a histogram. What geom could be the right one?" ) ) pass() })
Re-create this histogram from above, but set the width of the histogram bins
to 10.
ggplot( data = nycflights21_small, mapping = aes(x = arr_delay) ) + geom_histogram(____) + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x = arr_delay) ) + geom_histogram(____ = 10) + theme_bw()
ggplot( data = nycflights21_small, mapping = aes(x = arr_delay) ) + geom_histogram(binwidth = 10) + theme_bw()
grade_this({ fail_if(!ggcheck::is_ggplot(.result), message = paste0( "You did not define a `ggplot` object. Use the function ", "`ggplot2::ggplot()`. ", "Did you maybe forget to call and render the plot?" ) ) fail_if(!ggcheck::uses_data(.result, nycflights21_small), message = paste0( "You should use the data set `nycflights21_small`. You can pass ", "it to the argument `data` of `ggplot2::ggplot()`." ) ) fail_if(!uses_mappings(.result, aes(x=arr_delay), exact = TRUE), message = paste0( "There is a problem with your mapping. You only need one variable " , "this time. Think about which axis it would make sense to plot it ", "onto. As usual, use the argument `mapping` ", "of the `ggplot2::ggplot()` function, as well as `ggplot2::aes()`." ) ) fail_if(!ggcheck::uses_geoms(.result, "histogram", exact = FALSE), message = paste0( "You must produce a histogram. What geom could be the right one?" ) ) fail_if(!uses_geom_params(.result, "histogram", list(binwidth = 10)), message = paste0( "You forgot the set the binwidth of the histogram. It might be ", "a good idea to google how to set the binwidth (not the nb of bins!)!" ) ) pass() })
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.