Chapter 6 Data Visualization

6.1 ggplot2 () Package - Data Visualization Tool

The package ggplot2 () is a very powerful tool for data visualization.

Please use the following website to develop this chapter - https://r-graphics.org/index.html

6.2 Scatter Plot

In scatter Plot, we use to continuous variables in x axis and y axis. The following code is run to create a scatter plot of two variables called time_hour and total_delay. See Figure 6.1.

library(nycflights13)
library(tidyverse)
flights <- flights %>% 
  mutate(total_delay = dep_delay + arr_delay)

flights %>% 
  ggplot(mapping = aes( x = time_hour, y = total_delay))+
  geom_point()

## Warning: Removed 9430 rows containing missing values
## (geom_point).

FIGURE6.1: A Scatterplot of Total Delay and Month

However, we can also use a third variable in scatter plot. See Figure 6.2. Also see Figure 6.3.

flights %>% 
  ggplot(mapping = aes( x = time_hour, y = total_delay))+
  geom_point()+
  facet_wrap(~origin)

## Warning: Removed 9430 rows containing missing values
## (geom_point).

FIGURE6.2: A Scatterplot of Total Delay and Month in Departing Airports in New York

flights %>% 
  ggplot(mapping = aes( time_hour, total_delay, color = origin))+
  geom_point()

## Warning: Removed 9430 rows containing missing values
## (geom_point).

FIGURE6.3: A Scatterplot of Total Delay and Month Using Origin as Third Variable

6.3 Line Plot

In line plot, we draw line for the points of two continuous variables. See Figure 6.4.

flights %>% 
  group_by(month) %>% 
  summarize(avg_delay = mean (total_delay, na.rm = TRUE)) %>%
  ggplot(mapping = aes(x = month, y = avg_delay)) + 
  geom_smooth()

## `summarise()` ungrouping output (override with `.groups` argument)

## `geom_smooth()` using method = 'loess' and formula 'y ~ x'

FIGURE6.4: A lineplot of Average Delay and Month

Like scatter plot, a third variable can also be added to the line plot. See Figure 6.5. Also see Figure 6.6.

flights %>% 
  ggplot(mapping = aes(x = month, y = total_delay, color = origin)) + 
  geom_smooth()

## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'

## Warning: Removed 9430 rows containing non-finite values
## (stat_smooth).

FIGURE6.5: A lineplot of Average Delay and Month Using Origin as Third Variable

flights %>% 
  ggplot(mapping = aes(x = day, y = total_delay, color = origin)) + 
  geom_smooth()

## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'

## Warning: Removed 9430 rows containing non-finite values
## (stat_smooth).

FIGURE6.6: A lineplot of Total Delay and Day of the Week Using Origin as Third Variable

6.4 Bar Plot

We can also create bar diagram. See Figure 6.7.

ggplot(flights, aes(origin))+
  geom_bar()

FIGURE6.7: A barplot of Number of Flights from Departing Airports in New York

We can also include a second variable in bar diagram. Please see Figure 6.8. Also see Figure 6.9.

ggplot(flights, aes(origin))+
  geom_bar(aes(fill = as.character(month)))

FIGURE6.8: A barplot of Number of Flights from Departing Airports in New York in Different Months

ggplot(flights, aes(origin))+
  geom_bar(aes(fill = carrier))

FIGURE6.9: A barplot of Number of Flights from Departing Airports in New York by Different Carriers

From the graph, we can figure out which carriers use which airport most.

6.5 Box Plot

We can also create boxplot using ggplot2. Please see Figure 6.10.

ggplot(flights, aes ( origin,distance))+
  geom_boxplot()

FIGURE6.10: A boxplot of Distance from Departing Airports in New York

It is evident that long distance flights use JFK because it has the highest distance. A thrid variable also can be included in box plot. Please see Figure 6.11.

ggplot(flights, aes ( origin,distance))+
  geom_boxplot()+
  facet_wrap(~ month)

FIGURE6.11: A boxplot of Distance from Departing Airports in New York in Different Months

Here boxplot is created for the distance variables by origin and month variables