ggplot-sol
Jose Rojas
2025-07-16
Solutions
us_data <- filter(gapminder, country == "United States")Q01. Replicate that scatterplot using ggplot() with a points layer
ggplot(data = us_data, aes(x = gdpPercap, y = lifeExp)) +
geom_point()
Q02. Add a main title and axis labels (labs) to your ggplot
us_data %>%
ggplot(data = ., aes(x = gdpPercap, y = lifeExp)) +
geom_point() +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "GDP Per Capita vs Life Exp. in US")
Q03. Use multiple geom’s in a ggplot (if there’s no blank ’__’, you don’t need to write anything!):
us_data %>%
ggplot(data = ., aes(x = gdpPercap, y = lifeExp)) +
geom_point() +
geom_line() +
geom_smooth()## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
Q04. Change the plot above to facet by ‘year’ instead of faceting by ‘continent’.
ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp)) +
geom_point() +
facet_wrap(~ year)
Q05. Draw a scatterplot comparing gdpPercap and lifeExp, where different continents are drawn with different colors.
ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp, color = continent)) +
geom_point()
Q06. Use ‘color’ to represent ‘continent’, and use ‘fill’ to represent ‘year’.
ggplot(
data = gapminder,
aes(x = gdpPercap, y = lifeExp, color = continent, fill = year)
) +
geom_point(shape = 21)
Q07. Try mapping ‘shape’ to ‘continent’ inside of an aes() call.
ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp)) +
geom_point(aes(shape = continent))
Q08. So far we have 3 aesthetic mappings other than x and y axis:
They are ‘color’, ‘fill’, and ‘shape’. The last ‘aes’ we’ll talk about is ‘size’, which adjusts point size. Add 2 ‘aes’ to this plot: map ‘continent’ to ‘color’, and map ‘pop’ to ‘size’. How many legends are there now?
ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp)) +
geom_point(aes(color = continent, size = pop))
Q09. Draw a histogram to visualize lifeExp using geom_histogram().
ggplot(
data = gapminder,
aes(x = lifeExp, color = continent, fill = continent)
) +
geom_histogram()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Q10. Experiment with different binwidths for your histogram.
ggplot(
data = gapminder,
aes(x = lifeExp, color = continent, fill = continent)
) +
geom_histogram(binwidth = 5)
Q11. geom_freqpoly() is just like a histogram, but it uses lines instead of bars to communicate the number of observations in each bin.
Again, experiment with binwidth.
ggplot(
data = gapminder, aes(x = lifeExp, color = continent)
) +
geom_freqpoly(binwidth = 10)
Q12. Make a filled frequency polygon using an area plot: geom_area().
In the blank, experiment with setting ‘color’ versus ‘fill’ as ‘continent’.
ggplot(data = gapminder, aes(x = lifeExp, color = continent)) +
geom_area(stat = "bin")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ggplot(data = gapminder, aes(x = lifeExp, fill = continent)) +
geom_area(stat = "bin")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Q13. Notice that the continents are stacked up on top of each other in the previous plot.
To change that behavior, set position = “dodge”.
ggplot(data = gapminder, aes(x = lifeExp, fill = continent)) +
geom_area(stat = "bin", position = 'dodge')## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Q15. Experiment with setting ‘color’ and ‘fill’ to ‘continent’.
ggplot(data = gapminder, aes(x = lifeExp, fill = continent)) +
geom_density()
ggplot(data = gapminder, aes(x = lifeExp, color = continent)) +
geom_density()
Q16. With geom_density and fill, experiment with different alpha’s.
ggplot(data = gapminder, aes(x = lifeExp, fill = continent)) +
geom_density(alpha = 0.6)
Q17. Add a vertical line to this plot.
Hint: for a vertical line, you’ll need to specify an x-intercept (and for a horizontal line, you’d need to specify a y-intercept). Do this with ‘xintercept = 70’.
ggplot(
data = gapminder,
aes(x = lifeExp, color = continent, fill = continent)
) +
geom_density(alpha = .3) +
geom_vline(xintercept = 70)
Q18. Include the vertical line you did above, and also note how the annotation (‘annotate()’) labels the line.
ggplot(
data = gapminder,
aes(x = lifeExp, color = continent, fill = continent)
) +
geom_density(alpha = .3) +
geom_vline(xintercept = 70) +
annotate(
"text", x = 70, y = .075, label = "70 years", angle = 90
)
Q19. We’ve already talked about basic scatterplots. Try to recall how to draw a ggplot that plots ‘gdpPercap’ on the x-axis and ‘lifeExp’ on the y-axis.
ggplot(
data = gapminder,
aes(x = gdpPercap, y = lifeExp)
) +
geom_point()
Q20. Add a smoothed line layer to the plot you did above using geom_smooth().
ggplot(
data = gapminder,
aes(x = gdpPercap, y = lifeExp)
) +
geom_point() +
geom_smooth()## `geom_smooth()` using method = 'gam' and formula = 'y ~ s(x, bs = "cs")'
Q21. Remove the standard error ribbon using geom_smooth(se = FALSE).
ggplot(
data = gapminder,
aes(x = gdpPercap, y = lifeExp)
) +
geom_point(se = FALSE)## Warning in geom_point(se = FALSE): Ignoring unknown parameters: `se`
Q22. Visualize the OLS fit by using geom_smooth(method = “lm”).
“lm” stands for “linear model”.
ggplot(
data = gapminder,
aes(x = gdpPercap, y = lifeExp)
) +
geom_point() +
geom_smooth(method = "lm")## `geom_smooth()` using formula = 'y ~ x'
Q23. Instead of plotting ‘gdpPercap’ on the x-axis, plot the log of ‘gdpPercap’ using log().
You should find that a linear model seems to fit this transformation much better.
ggplot(data = gapminder, aes(x = log(gdpPercap), y = lifeExp)) +
geom_point() +
geom_smooth(method = "lm")## `geom_smooth()` using formula = 'y ~ x'
Notice that the units on the x-axis of the previous plot are in log terms, which is hard to interpret. I prefer this method: Do a log transformation of the x-axis using ‘scale_x_log10()’, and use ‘labels = scales::comma’ to suppress scientific notation on the labels for the x-axis.
ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp)) +
scale_x_log10(labels = scales::comma) +
geom_point() +
geom_smooth(method = "lm")## `geom_smooth()` using formula = 'y ~ x'
Q24. The second issue is that there’s over-plotting near the center.
I’d like a way to visualize how dense the points are getting. For this, we can replace ‘geom_point()’ with ‘geom_hex()’.
ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp)) +
scale_x_log10(labels = scales::comma) +
geom_hex() +
geom_smooth(method = "lm")## `geom_smooth()` using formula = 'y ~ x'
Q25. Draw a line plot using ‘geom_line’. Compare US ‘gdpPercap’ over time to two other countries of your choice.
gapminder %>%
filter(country %in% c("United States", "Costa Rica", "Mexico")) %>%
ggplot(aes(x = year, y = gdpPercap, color = country)) +
geom_line()
Q26. Draw a boxplot with geom_boxplot() that compares the gdpPercap (on the y-axis) of different continents (on the x-axis, also using color).
Experiment with applying a log transformation to gdpPercap (now on the y axis).
ggplot(
data = gapminder,
aes(x = continent, y = gdpPercap, color = continent)
) +
geom_boxplot()
ggplot(
data = gapminder,
aes(x = continent, y = log(gdpPercap), color = continent)
) +
geom_boxplot()
Q27. Change the plot from the previous question to be a violin plot using ‘geom_violin’.
Also include ‘fill’. Make sure you use ‘scale_y_log10()’ to transform the y-axis into log terms.
ggplot(
data = gapminder,
aes(x = continent, y = log(gdpPercap), color = continent, fill = continent)
) +
scale_y_log10() +
geom_violin()
Finally, check out how simple it is to change your ‘facet_wrap’ to a slick animation with ‘gganimate’!
The ‘facet_wrap’ version:
ggplot(
data = gapminder,
aes(x = log(gdpPercap), y = lifeExp)
) +
geom_point(aes(color = continent)) +
geom_density2d(color = "grey", alpha = .5) +
facet_wrap(~ year)