Data Visualization

University of St.Gallen

Gilbert Fontana

October 27, 2022

Outline

• About me 👋
• How I work with ggplot2 💻
  • Walkthrough of a plot 📊
• Questions and discussions 💭

About me

Education 🎓

• Bsc in Political Science
• Bsc in Economics
• Msc in Economics & Econometrics

Experience 🛠️

• Previously
  • Analyst at Statistic Sweden
  • Analyst at Kommuninvest
• Currently
  • Statistical Programmer at Saco
  • Creator at Visual Capitalist

#TidyTuesday

• A weekly data project by the
  R4DS Online Learning Community
  and Thomas Mock

• A great way to practice data
  visualization and learn from
  others!

Basic theory of ggplot

ggplot is based on the foundations in Leland Wilkinson’s book: The Grammar of Graphics.

It is built around the following principles:

• Data
• Aesthetics aes()
• Statistics stat_
• Scales scale_
• Geometries geom_
• Facets facet_
• Coordinates coord_
• Themes theme_

Visualization walkthrough

Visualization walkthrough

Packages 📦

#### LIBS ####
library(tidyverse)
library(showtext)
library(MetBrewer)

• tidyverse
  • ggplot2
  • dplyr
  • tidyr
  • readr
  • purrr
  • tibble
  • stringr
  • forcats

• showtext
  • A package for handling fonts

• MetBrewer
  • Palettes inspired by works at the Metropolitan Museum of Art in New York

Visualization walkthrough

Data 🗃️

#### DATA ####
df <- read_csv("https://raw.githubusercontent.com/gilbertfontana/External/main/University%20of%20St.%20Gallen/tax_data.csv")

country	year	centaxgdp
Denmark	1820	4
Denmark	1821	5
Denmark	1822	5
Denmark	1823	5
Denmark	1824	5

Visualization walkthrough

Style 🎨

#### STYLE ####
font <- "Oswald"
font_add_google(family=font, font, db_cache = FALSE)
showtext_auto(enable = TRUE) 
theme_set(theme_minimal(base_family = font))
bg <- "#334756"
txt_col <- "#DEA2A2"

Visualization walkthrough

Step 1

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line()

Visualization walkthrough

Step 2

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country))

Visualization walkthrough

Step 3

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(aes(color=country))

Visualization walkthrough

Step 4

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country))

Visualization walkthrough

Step 5

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  theme(legend.position = "none")

Visualization walkthrough

Step 6

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(aes(label=country)) +
  theme(legend.position = "none")

Visualization walkthrough

Step 7

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(data = df %>% filter(year=="2020"),
            aes(label=country)) +
  theme(legend.position = "none")

Visualization walkthrough

Step 8

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(data = df %>% filter(year=="2020"),
            aes(label=country,
                color=country),
            hjust = -.2,
            family = font,
            size = 4) +
  theme(legend.position = "none")

Visualization walkthrough

Step 9

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(data = df %>% filter(year=="2020"),
            aes(label=country,
                color=country),
            hjust = -.2,
            family=font,
            size = 4) +
  coord_cartesian(clip = "off") +
  theme(legend.position = "none",
        plot.margin = margin(30, 30, 30, 30))

Visualization walkthrough

Step 10

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(data = df %>% filter(year=="2020"),
            aes(label=country,
                color=country),
            hjust = -.2,
            family = font,
            size = 4) +
  coord_cartesian(clip = "off") +
  scale_color_manual(
    values = met.brewer(name = "Paquin",
                        type = "discrete",
                        n = 4)) +
  theme(legend.position = "none",
        plot.margin = margin(30, 30, 30, 30))

Visualization walkthrough

Step 11

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(data = df %>% filter(year=="2020"),
            aes(label=country,
                color=country),
            hjust = -.2,
            family = font,
            size = 4) +
  coord_cartesian(clip = "off") +
  scale_color_manual(
    values = met.brewer(name = "Paquin",
                        type = "discrete",
                        n = 4)) +
  scale_y_continuous(limits = c(0,40),
                     expand = c(0,0)) +
  theme(legend.position = "none",
        plot.margin = margin(30, 30, 30, 30))

Visualization walkthrough

Step 12

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(data = df %>% filter(year=="2020"),
            aes(label=country,
                color=country),
            hjust = -.2,
            family = font,
            size = 4) +
  coord_cartesian(clip = "off") +
  scale_color_manual(
    values = met.brewer(name = "Paquin",
                        type = "discrete",
                        n = 4)) +
  scale_y_continuous(limits = c(0,40),
                     expand = c(0,0)) +
  theme(
    panel.grid = element_blank(),
    axis.text = element_text(color = txt_col,
                             size = 10),
    axis.title = element_text(color = txt_col,
                              size = 12),
    legend.position = "none",
    plot.margin = margin(30, 30, 30, 30),
    plot.background = element_rect(fill = bg,
                                   color = bg)
  )

Visualization walkthrough

Step 13

df %>% 
  ggplot(aes(x=year,
             y=centaxgdp,
             group=country)) +
  geom_line(aes(color=country)) +
  geom_point(data = df %>% filter(year=="2020"),
             aes(color=country)) +
  geom_text(data = df %>% filter(year=="2020"),
            aes(label=country,
                color=country),
            hjust = -.2,
            family = font,
            size = 4) +
  coord_cartesian(clip = "off") +
  scale_color_manual(
    values = met.brewer(name = "Paquin",
                        type = "discrete",
                        n= 4 )) +
  scale_y_continuous(limits = c(0,40),
                     expand = c(0,0)) +
  labs(title = "The Rise of the\nNordic Welfare States",
       caption = "Gilbert Fontana | Data: Andersson & Brambor (2019) and Eurostat",
       x = "",
       y = "Central government tax revenue as a share of GDP (%)") +
  theme(
    panel.grid = element_blank(),
    axis.text = element_text(color = txt_col,
                             size = 10),
    axis.title = element_text(color = txt_col,
                              size = 12),
    plot.title = element_text(size = 36,
                              hjust =.5,
                              color = txt_col,
                              face = "bold",
                              margin = margin(0,0,30,0)),
    plot.caption = element_text(hjust = .5,
                                color= txt_col),
    legend.position = "none",
    plot.margin = margin(30, 30, 30, 30),
    plot.background = element_rect(fill = bg,
                                   color = bg)
  )

Visualization walkthrough

Speaker notes go here.

Thank you!