Heatmaps

Graph info

Should I use this graph?

This graph requires:

✅ two numeric (continuous) variables

✅ a categorical variable

Description

Heatmaps display quantitative values across an intersection of two categorical (or discrete) variables.

The intersecting cells contain variations of color saturation (i.e., the grade of purity or vividness) to represent the numerical values between groups.

Heatmap legends should be positioned on top or bottom and justified horizontally to preserve shape and improve readability.

Getting set up

Packages

PACKAGES:

Install packages.

Code

install.packages("fivethirtyeight")
library(fivethirtyeight) 
library(ggplot2)

Data

DATA:

For the heatmap, we’re going to re-structure and filter the bob_ross data from the fivethirtyeight package.

Code

heatmap_ross <- fivethirtyeight::bob_ross |>
  pivot_longer(-c(episode, season,
    episode_num, title),
    names_to = "object",
    values_to = "present") |>
  mutate(present = as.logical(present),
    object = str_replace_all(object, "_", " ")) |>
  arrange(episode, object) |> 
  filter(object %in% c("conifer", "trees", 
    "tree", "snow", "palm trees", "grass", 
    "flowers", "cactus", "bushes", "cirrus", 
    "cumulus", "deciduous", "clouds", "fog")) |>
  group_by(season, object) |>
  summarise(occurrences = sum(present)) |>
  ungroup()

The grammar

Code

CODE:

Create labels with labs()

Initialize the graph with ggplot() and provide data

Assign season to x, object to y, and occurrences to fill

Add the geom_tile()

Move the legend to the bottom with theme(legend.position = "bottom")

Code

labs_heatmap_tile <- labs(
  title = "Bob Ross' plants & clouds", 
  x = "Episode", 
  y = "Plant & Cloud Objects", 
  fill = "Occurrences")
ggp2_heatmap_tile <- ggplot(data = heatmap_ross, 
       aes(x = season, 
           y = object, 
           fill = occurrences)) + 
       geom_tile() + 
    theme(legend.position = "bottom")
ggp2_heatmap_tile + 
  labs_heatmap_tile

Graph

GRAPH:

More info

In addition to geom_tile(), heatmaps can also be created with the geom_raster() function.

geom_raster()

geom_raster():

Create labels with labs()

Initialize the graph with ggplot() and provide data

Assign season to x, object to y, and occurrences to fill

Add the geom_raster()

Move the legend to the bottom with theme(legend.position = "bottom")

Code

labs_heatmap_raster <- labs(
  title = "Bob Ross' plants & clouds", 
  x = "Episode", 
  y = "Plant/Cloud Object", 
  fill = "Occurrences")
ggp2_heatmap_raster <- ggplot(data = heatmap_ross, 
       aes(x = season, 
           y = object, 
           fill = occurrences)) + 
       geom_raster() + 
    theme(legend.position = "bottom")
ggp2_heatmap_raster + 
  labs_heatmap_raster