In hirscheylab/tidybiology: Utility Functions and Data Sets for Tidy Biology

knitr::opts_chunk$set(echo = FALSE)
# Load packages

# install.packages("tidybiology")
library(tidyverse)
# install.packages("ggExtra")
library(ggExtra)
library(tidybiology)

# Load data
data(happy_full)
data(happy_select)

Visualizing data from the World Happiness Report

Now we move to the fun part - making pretty plots!

Plotting distributions

Let's begin by getting a sense of the overall distribution of ladder_score in the happy_select data frame

happy_select %>% 
  ggplot(aes(ladder_score)) +
  geom_histogram()

Making improvements

This plot is fine, but it's a little chunky. Let's try a different geom - geom_density() - to see what we get

happy_select %>% 
  ggplot(aes(ladder_score)) +
  geom_density()

Density by region

What does the ladder_score distribution look like for each region? There are many ways to visualize this. Let's first plot all the distributions on one plot

happy_select %>% 
  ggplot(aes(ladder_score, fill = region)) +
  geom_density()

Nice! One small problem with this plot is that distributions are overlapping, making it difficult to visualize everything. Which argument can you adjust to improve this plot? Does it go within aes() or outside? Why?

Faceting

Let's consider an alternative way to plot the distributions. Go ahead and use the facet_wrap() function to do this

happy_select %>% 
  ggplot(aes(ladder_score)) +
  geom_density() +
  facet_wrap(~region)

Scatterplots

Are rich countries "happier"?

We can use scatterplots to visualize the relationship between two variables. For example, let's take a look at the relationship between ladder_score and logged_gdp_per_capita (this relationship might be obvious, but maybe we'll be surprised!)

happy_select %>% 
  ggplot(aes(logged_gdp_per_capita, ladder_score)) +
  geom_point()

Emphasizing relationships

It's clear from the previous scatterplot that relationship between ladder_score and logged_gdp_per_capita is positive, and it looks pretty linear. To highlight this, we can use the geom_smooth() geom (remember to set the se argument to equal FALSE to just get a straight line)

happy_select %>% 
  ggplot(aes(ladder_score, logged_gdp_per_capita)) +
  geom_point() +
  geom_smooth(se = FALSE)

Making things prettier

The plots we've made, while nice, look a little...basic. Let's try to make some improvements. We'll use the last plot

Add a title, subtitle, x-axis label, and y-axis label. Also change the overall look of the plot using your favourite theme_*() function

happy_select %>% 
  ggplot(aes(ladder_score, logged_gdp_per_capita)) +
  geom_point() +
  geom_smooth(se = FALSE) +
  labs(title = "Does Money == Happiness?",
       subtitle = "Plotting Ladder Score against Log GDP Per Capita",
       x = "Ladder Score",
       y = "Log GDP Per Capita") +
  theme_minimal()

Adding marginal plots

Add histograms for logged_gdp_per_capita and ladder_score to the previous plot (to the top and to the right, respectively)

Hint: Use the ggMarginal() function from the ggExtra package. To learn more about this function, run ?ggMarginal

p1 <- happy_select %>% 
        ggplot(aes(ladder_score, logged_gdp_per_capita)) +
        geom_point() +
        geom_smooth(se = FALSE) +
        labs(title = "Does Money == Happiness?",
             subtitle = "Plotting Ladder Score against Log GDP Per Capita",
             x = "Ladder Score",
             y = "Log GDP Per Capita") +
        theme_minimal()

ggMarginal(
  p = p1,
  type = 'histogram'
)

Applying the "final" touch

Make the x- and y-axis text bold

p1 <- happy_select %>% 
        ggplot(aes(ladder_score, logged_gdp_per_capita)) +
        geom_point() +
        geom_smooth(se = FALSE) +
        labs(title = "Does Money == Happiness?",
             subtitle = "Plotting Ladder Score against Log GDP Per Capita",
             x = "Ladder Score",
             y = "Log GDP Per Capita") +
        theme_minimal() +
        theme(axis.text = element_text(face = "bold"))

ggMarginal(
  p = p1,
  type = 'histogram'
)

Save the last plot

ggsave(file = "my_plot.png", ggMarginal(p = p1, type = 'histogram'))

For more help

Run the following to access the Ggplot2 vignette

browseVignettes("ggplot2")

hirscheylab/tidybiology documentation built on May 20, 2022, 10:55 p.m.