Intro to comparing distributions

Visualization Best Practices in R

Nick Strayer

Instructor

Why compare distributions?

Verify balanced groups
For comparison's sake

Why not facet histogams?

ggplot(md_speeding, aes(x = speed_over)) + 
  geom_histogram() +
  facet_grid(vehicle_color ~ .)

The boxplot

Boxplot pros

Familiar
Lots of good summary statistics

boxplot cons

Show me the data!

A simple addition

geom_jitter() shows raw points jostled to avoid overlap.
Layer under your geom_boxplot().

md_speeding %>% 
  filter(vehicle_color == 'BLUE') %>%
  ggplot(aes(x = gender, y = speed)) +
    # Draw points behind 
    geom_jitter(alpha = 0.3, color = 'steelblue') + 
    # Make transparent
    geom_boxplot(alpha = 0) + 
    labs(title = 'Distribution of speed for blue cars by gender')

Let's compare some distributions!

Visualization Best Practices in R

Preparing Video For Download...