Import your data

artists <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-09-27/artists.csv') 
## Rows: 3380 Columns: 7
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (3): state, race, type
## dbl (4): all_workers_n, artists_n, artists_share, location_quotient
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Chapter 15

Create a factor

race <- c("Hispanic", "White", "African-American", "Asian", "Other")
race
## [1] "Hispanic"         "White"            "African-American" "Asian"           
## [5] "Other"
race_factor <- factor(race, levels = race)
race_factor
## [1] Hispanic         White            African-American Asian           
## [5] Other           
## Levels: Hispanic White African-American Asian Other

Modify factor order

Make two bar charts here - one before ordering another after

Unordered factor levels

# Transform data: group by race, average artists by race 
avg_artists <- artists %>%
    
    group_by(race) %>%
    summarise(avg_artists = mean(artists_n, na.rm = TRUE))

# Plot
avg_artists %>%
    ggplot(aes(x = avg_artists, y = race)) +
    geom_point()

Ordered factor levels

avg_artists %>%
    ggplot(aes(x = avg_artists, y = fct_reorder(.f = race, .x = avg_artists))) +
    geom_point() +
    labs(y = NULL, x = "Average Artists by Race")

Modify factor levels

Show examples of three functions:

  • fct_recode
artists %>% distinct(race)
## # A tibble: 5 × 1
##   race            
##   <chr>           
## 1 Hispanic        
## 2 White           
## 3 African-American
## 4 Asian           
## 5 Other
# Recode function
artists %>%
#Rename levels
    mutate(race_rev = fct_recode(race, "Caucasian" = "White")) %>%
    select(race, race_rev) %>%
    filter(race == "White")
## # A tibble: 676 × 2
##    race  race_rev 
##    <chr> <fct>    
##  1 White Caucasian
##  2 White Caucasian
##  3 White Caucasian
##  4 White Caucasian
##  5 White Caucasian
##  6 White Caucasian
##  7 White Caucasian
##  8 White Caucasian
##  9 White Caucasian
## 10 White Caucasian
## # ℹ 666 more rows
# Collapse multiple levels into one
artists %>%
    mutate(race_col = fct_collapse(race, "Minority" = c("African-American", "Other"))) %>%
    select(race, race_col) %>%
    filter(race != "White") %>%
    filter(race != "Hispanic") %>%
    sample_n(10)
## # A tibble: 10 × 2
##    race             race_col
##    <chr>            <fct>   
##  1 Other            Minority
##  2 Asian            Asian   
##  3 African-American Minority
##  4 Other            Minority
##  5 Other            Minority
##  6 African-American Minority
##  7 African-American Minority
##  8 African-American Minority
##  9 Asian            Asian   
## 10 Asian            Asian
# Lump small levels into other levels
artists %>% count(race)
## # A tibble: 5 × 2
##   race                 n
##   <chr>            <int>
## 1 African-American   676
## 2 Asian              676
## 3 Hispanic           676
## 4 Other              676
## 5 White              676
artists %>% 
   mutate(race_lump = fct_lump(race, n = 3)) %>%
    count(race_lump, sort = TRUE)
## # A tibble: 5 × 2
##   race_lump            n
##   <fct>            <int>
## 1 African-American   676
## 2 Asian              676
## 3 Hispanic           676
## 4 Other              676
## 5 White              676
# Because there are no minimum, maximum or unique values for each of the race levels, the race_lump function does not lump.

Chapter 16

No need to do anything here.