DataViz: Polishing
Heather Kitada Smalley
Adapted from a lab written by Prof Charlotte Wickham
The data
These data come from the hflights package, but some
summarization is done for you.
library(tidyverse)
iah <- read.csv("https://raw.githubusercontent.com/kitadasmalley/Teaching/main/DATA502/FA2023/R_Markdown/Week8/iah_flightSummary.csv")
str(iah)## 'data.frame': 156 obs. of 6 variables:
## $ DayOfWeekNum : int 1 1 1 1 1 1 1 1 1 1 ...
## $ DepHour : int 0 1 5 6 7 8 9 10 11 12 ...
## $ prop_over_15 : num 1 1 0.00272 0.03348 0.03315 ...
## $ avg_delay : num 188.818 526.333 -3.147 1.206 0.226 ...
## $ DayOfWeek : chr "Mon" "Mon" "Mon" "Mon" ...
## $ avg_delay_delayed: num 188.82 526.33 6.25 10.76 7.15 ...Start with this heatmap:
The heatmap represents the proportion of flights that have a
departure delay of more that 15 minutes (prop_over_15) at
the George Bush Intercontinental Airport (IAH) by day of the week
(DayOfWeek) and departure hour (DepHour).
# make sure days of week are displayed in the right order
iah$DayOfWeek <- factor(iah$DayOfWeek,
levels = c("Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"))
p <- ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15))
p## Warning: Removed 7 rows containing missing values (geom_tile).
Step 1 : scale_x_continuous
Breaks
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24)) ## Warning: Removed 7 rows containing missing values (geom_tile).
Labels
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight")) ## Warning: Removed 7 rows containing missing values (geom_tile).
#### Limits
0 and midnight are the same thing!
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
limits = c(0, 23)) ## Warning: Removed 8 rows containing missing values (geom_tile).
Expansion (Padding Space)
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23))## Warning: Removed 8 rows containing missing values (geom_tile).
Step 2: scale_y_discrete
Label
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") ## Warning: Removed 8 rows containing missing values (geom_tile).
Step 3: scale_fill_continuous_sequential
We’re going to need the color space package for this.
Palette
library(colorspace)
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") +
scale_fill_continuous_sequential(
name = "Flights delayed \nmore than 15 mins",
palette = "YlOrRd")## Warning: Removed 8 rows containing missing values (geom_tile).
Breaks
library(colorspace)
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") +
scale_fill_continuous_sequential(
name = "Flights delayed \nmore than 15 mins",
palette = "YlOrRd",
breaks = c(0, .25, 0.5, .75, 1))## Warning: Removed 8 rows containing missing values (geom_tile).
Labels
library(colorspace)
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") +
scale_fill_continuous_sequential(
name = "Flights delayed \nmore than 15 mins",
palette = "YlOrRd",
breaks = c(0, .25, 0.5, .75, 1),
labels = c("0%", "25%", "50%", "75%", "100%"))## Warning: Removed 8 rows containing missing values (geom_tile).
Guide (Legend)
library(colorspace)
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") +
scale_fill_continuous_sequential(
name = "Flights delayed \nmore than 15 mins",
palette = "YlOrRd",
breaks = c(0, .25, 0.5, .75, 1),
labels = c("0%", "25%", "50%", "75%", "100%"),
expand = c(0, 0),
guide = "colorbar")## Warning: Removed 8 rows containing missing values (geom_tile).
Step 4: Title and subtitle
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") +
scale_fill_continuous_sequential(
name = "Flights delayed \nmore than 15 mins",
palette = "YlOrRd",
breaks = c(0, .25, 0.5, .75, 1),
labels = c("0%", "25%", "50%", "75%", "100%"),
expand = c(0, 0),
guide = "colorbar") +
labs(title = "Take an early flight from IAH to avoid delays",
subtitle = "Based on all departing flights from George Bush Intercontinental Airport (IAH) in 2011")## Warning: Removed 8 rows containing missing values (geom_tile).
Step 5: Theme
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") +
scale_fill_continuous_sequential(
name = "Flights delayed \nmore than 15 mins",
palette = "YlOrRd",
breaks = c(0, .25, 0.5, .75, 1),
labels = c("0%", "25%", "50%", "75%", "100%"),
expand = c(0, 0),
guide = "colorbar") +
labs(title = "Take an early flight from IAH to avoid delays",
subtitle = "Based on all departing flights from George Bush Intercontinental Airport (IAH) in 2011") +
theme_classic() +
theme(axis.ticks.y = element_blank(), axis.line.y = element_blank()) + coord_equal()## Warning: Removed 8 rows containing missing values (geom_tile).
Step 6: Order Days (Relevel)
iah$DayOfWeek <- factor(iah$DayOfWeek,
levels = c("Sun", "Sat", "Fri", "Thu", "Wed", "Tue","Mon"))
ggplot(iah, aes(DepHour, DayOfWeek)) +
geom_tile(aes(fill = prop_over_15), colour = "grey50") +
scale_x_continuous("Departure time",
breaks = c(0, 6, 12, 18, 24),
labels = c("midnight", "6am", "noon", "6pm", "midnight"),
expand = c(0, 0),
limits = c(0, 23)) +
scale_y_discrete("Departure day") +
scale_fill_continuous_sequential(
name = "Flights delayed \nmore than 15 mins",
palette = "YlOrRd",
breaks = c(0, .25, 0.5, .75, 1),
labels = c("0%", "25%", "50%", "75%", "100%"),
expand = c(0, 0),
guide = "colorbar") +
labs(title = "Take an early flight from IAH to avoid delays",
subtitle = "Based on all departing flights from George Bush Intercontinental Airport (IAH) in 2011") +
theme_classic() +
theme(axis.ticks.y = element_blank(), axis.line.y = element_blank()) + coord_equal()## Warning: Removed 8 rows containing missing values (geom_tile).
Step 7: ggsave()
ggsave("polishedDelays.png", height = 3.5, width = 10)## Warning: Removed 8 rows containing missing values (geom_tile).