Data 607 Week 5

Wrangling of the Flights Data

load in the neccssary Libraries

library(stringr)
library(knitr)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(ggplot2)
library(tidyr)

Construct the Initial Dataset

#get the data from my github page
Raw_text <- read.csv("https://raw.githubusercontent.com/crarnouts/CUNY-MSDS/master/Arrival_times_2.csv", header = TRUE)
#turn all of the blanks into N/A or missing values
is.na(Raw_text) <- Raw_text==''
flights<-Raw_text
#name the columns of the dataset 
colnames(flights)[1:2] = c("Airline", "Status")
#fill in the blanks of the airline column
flights <- flights %>% fill(Airline)

Summarize the proportion of flights by the airline and by whether or not the flight was on time

flights.tidy <- gather(flights,"City","FlightCount",-Airline,-Status)
#Summarize the number of flights by the airline
flights.total <- flights.tidy %>% 
  group_by(Airline) %>% 
  summarise(flights = sum(FlightCount))
#summarize the number of flights by the status of the flight
flights.time <- flights.tidy %>% 
  group_by(Status) %>% 
  summarise(flights = sum(FlightCount))
#Create proportions based off the above criteria
flights.time$proportion <- flights.total$flights / sum(flights.total$flights)
flights.total$proportion <- flights.total$flights / sum(flights.total$flights)

Graphic Visualizations of the data

#Graph the total flight count by airline
ggplot(flights.total,aes(x=Airline,y=flights)) + 
  geom_bar(stat="identity",color = "Yellow", fill = "Yellow") +
  ggtitle("Flight Count by Airline") +
  theme(plot.title = element_text(hjust = 0.5))

#graph the status count of the flights
ggplot(flights.time,aes(x=Status,y=flights)) + 
  geom_bar(stat="identity",color = "Orange", fill = "Orange") +
  ggtitle("Flight Count by Status") +
  theme(plot.title = element_text(hjust = 0.5))

#use the pipe operator to determine the total number of delayed flights by airline
flights.delayed <- flights.tidy %>% 
  group_by(Airline,Status) %>% 
summarise(flights = sum(FlightCount)) %>% 
filter(Status == "delayed ")
#use the flights.delayed dataset to determine the proportion of on time and delayed flights by airline
flights.total$delayed <- flights.delayed$flights
flights.total$delayed_proportion <- flights.total$delayed / flights.total$flights
flights.total$ontime_proportion <- 1 - flights.total$delayed_proportion

#Graph the ontime proporation by airline
ggplot(flights.total,aes(x=Airline,y=ontime_proportion)) + 
  geom_bar(stat="identity",fill="Red") +
  ggtitle("Proportion of On Time Flights by Airline ") +
  ylab("proportion") +
  theme(plot.title = element_text(hjust = 0.5))