Import Data

# CSV file
artists <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-09-27/artists.csv')

Introduction

Questions

Variation

Visualizing distributions

# Total artists per race
ggplot(data = artists) +
  geom_bar(mapping = aes(x = type))

artists %>% count(type)
## # A tibble: 13 × 2
##    type                                           n
##    <chr>                                      <int>
##  1 Actors                                       260
##  2 Announcers                                   260
##  3 Architects                                   260
##  4 Dancers And Choreographers                   260
##  5 Designers                                    260
##  6 Entertainers                                 260
##  7 Fine Artists, Art Directors, And Animators   260
##  8 Landscape Architects                         260
##  9 Music Directors And Composers                260
## 10 Musicians                                    260
## 11 Photographers                                260
## 12 Producers And Directors                      260
## 13 Writers And Authors                          260
# Percentage of artists in U.S. population
ggplot(data = artists) +
  geom_histogram(mapping = aes(x = artists_share))

# Total artists by race
ggplot(data = artists, mapping = aes(x = artists_n, colour = race)) +
  geom_freqpoly()

Typical values

ggplot(data = artists, mapping = aes(x = artists_share)) +
  geom_histogram()

Unusual values

# Outliers in percentage of U.S. artists
ggplot(artists) + 
  geom_histogram(mapping = aes(x = artists_share)) +
  coord_cartesian(xlim = c(0, 0.013))

Missing Values

# Total employed as artists per occupation type
ggplot(data = artists, mapping = aes(x = artists_n, y = type)) + 
  geom_point(na.rm = TRUE)

## Covariation

A categorical and continuous variable

# Artist occupation percentage by state
ggplot(data = artists, mapping = aes(x = artists_share, y = state)) +
  geom_boxplot()

Two categorical variables

# The proportion of artist type classified by artist race
artists %>% 
  count(race, type) %>%  
  ggplot(mapping = aes(x = race, y = type)) +
    geom_tile(mapping = aes(fill = n))

Two continous variables

# The relationship between all workers and all artists
ggplot(data = artists) +
  geom_point(mapping = aes(x = all_workers_n, y = artists_n)) +
  coord_cartesian(xlim = c(0, 500000), ylim = c(0, 80000))

Patterns and models