Week5

Airline CSV

What writes great csv files R can read without a problem? R does. So let us create a matrix (transposed because I misread things), and write that out. We then read it back in.

airlineData<-t(matrix(c("","","Los Angeles", "Phoenix", "San Diego", "San Francisco", "Seattle","Alaska", "on time", 497,221,212,503,1841,"","delayed",62,12,20,102,305,"","","","","","","","AM WEST","on time",694,4840,383,320,201,"","delayed",117,415,65,129,61),nrow=7))
write.csv(airlineData,"airlineData.csv")
airlineData <- read.csv("airlineData.csv", as.is = TRUE)
kable(airlineData)

Smashed Data.

First we take off the left index count and get a data frame. Then we take the top row and make the the names of the data. Next we cleanup the blank row. Then we dupe the category level labels.

airlineData<-as.data.frame(airlineData[-1],stringsAsFactors = FALSE)
names(airlineData)<-airlineData[1,]
names(airlineData)[1:2]<-c("Airline","Flight Status")
airlineData<-airlineData[c(-1,-4),]
airlineData[2,1]<-airlineData[1,1]
airlineData[4,1]<-airlineData[3,1]
kable(airlineData)

Because it says to use tidyr we can’t just use the base functions quixotically. So we gather into City with Flights excluding airlinge and flight status and sorting on airline.

airlineDataLong <- gather(airlineData ,key=City, value="Flights", -"Flight Status",-Airline) %>% arrange(Airline)
airlineDataLong$Flights<-as.numeric(airlineDataLong$Flights)
kable(airlineDataLong)

onTime <- airlineDataLong %>% spread(`Flight Status`,"Flights")
onTime$onTimePercent <- onTime$`on time` / (onTime$delayed+ onTime$`on time`) *100
onTime <-onTime %>% arrange(onTimePercent)
ggplot(onTime, aes(City, onTimePercent, color=Airline, size=40))+geom_point()