#Check to see if packages are already installed #if packages are not installed - install packages if (!require("rstudioapi")) { install.packages("rstudioapi") } if (!require("ggplot2")) { install.packages("ggplot2") } #import libraries library(rstudioapi) library(ggplot2) #get the current path of the working directory #this is to ensure that the data file and script are in the same folder path current_path = rstudioapi::getActiveDocumentContext()$path #set working directory to the current path setwd(dirname(current_path)) #Task 1 ##################################################################### # Task 1.1 # # - Check roller_coaster.csv exists in working directory # # - if file does not exist print warning and break to prevent # # to following script from executing # # - if file exists: # # - read csv into a dataframe named 'coasterData' # # - ensuring that headers are imported using 'header=T' # # - ensure that material_type, seating_type, manufacturer, # # and status are imported as factors # # - print 'coasterData' dataframe # ##################################################################### if(!file.exists("roller_coasters.csv")) { print("No CSV file found") break } else { coasterData <- data.frame(read.csv("roller_coasters.csv",header=T, colClasses=c(material_type="factor",seating_type="factor",manufacturer="factor",status="factor"))) coasterData } ##################################################################### # Task 1.2 # # - Create function that orders a dataframe by height and returns # # the top n records # # - create dataframe called 'heighestData' based on material_type # # - use top_n_records function to return highest coaster for each # # material_type # # - n = 1 denotes top record for each category # # - return columns name,height,park, and status using [] to # # specify the returned columns # # - return dataframe named 'highestDate' # ##################################################################### #Top N records function - input values = dataframe and the amount of records to be returned top_n_records <- function(df, n) { df[order(df$height, decreasing = TRUE), ][1:n, ] } #creating dataframe to return top record for each material_type by height highestData <- data.frame(do.call(rbind, by(coasterData, coasterData$material_type, top_n_records, n=1)))[c('name','height','park','status')] #return dataframe highestData ##################################################################### # Task 1.3 # # - create function that will display a bar chart showing the # # top 10 roller coasters, by speed, where the steating_type # # is 'inverted' # # - filter dataframe based on the seating_type that is passed # # into the function # # - filter dataframe to show the top 10 speeds # # - use ggplot to plot a box plot with speed as the X axis and # # name as the Y axis # # - style the box plot using the ggplot pre-defined functions # # - dynamically set the sub heading using the variable passed # # into the function # ##################################################################### #define function called 'displayComparison' with the dataframe and seating_type as variables displayComparison <- function(df,seating_type) { #filter dataframe based on seatring type df <- df[df$seating_type==seating_type,] #filer dataframe to top 10 decreasing to return the highest speeds df <- df[order(df$speed, decreasing = TRUE), ][1:10, ] #use ggplot to plot bar chart ggplot( #input dataframe data=df, #define the column names used for the box chart aes(x=speed,y=reorder(name,speed)) ) + # define style and colour theme of the boxplot geom_bar( stat="identity",fill="steelblue" )+ # style the labels and text colour and size geom_text( aes(label=speed),vjust=0.5, hjust=1.25, color="white", size=3.5 )+ # define label text with dynamic subtitle labs( title="Top 10 Roller Coasters by speed \n where the seat type is:", subtitle=seating_type, x="Speed", y="Roller Coaster" )+ # futher theme styling for labels and titles theme( plot.title = element_text(color="Black", size=12, face="bold", hjust=0.5), plot.subtitle = element_text(color="red",size=14,face="bold",hjust=0.5), axis.title.x = element_text(color="#993333", size=12), axis.title.y = element_text(color="#993333", size=12) ) } #run the function with coasterData and "inverted" as variables displayComparison(coasterData,"Inverted") ##################################################################### # Task 1.4 # # - create function 'correlateLengthSpeed' that takes a # # dataframe as a variable # # - filter dataframe where the type is equal to 'Sit Down' # # - return a dataframe the shows mean height and length and the # # correlation between of speed and height # # - ensure that there is no NA values in each field # ##################################################################### # define function that accepts a dataframe a a variable correlateLengthSpeed <- function(df) { #filter dataframe based on the type of coaster as 'Sit Down' sit_down_coasters <- df[df$seating_type == 'Sit Down',] #return dataframe return( data.frame ( # calculate the mean height based on the height column with no NA's Mean_Height = mean(sit_down_coasters[!is.na(sit_down_coasters$height),]$height), # calculate the mean length based on the length column with no NA's Mean_Length = mean(sit_down_coasters[!is.na(sit_down_coasters$length),]$length), # use cor function in R to calculate the correlation between length and speed # use 'pariwise.complete.obs' to remove any NA pairs within the column values Correlation_Length_Speed = cor(sit_down_coasters$length, sit_down_coasters$speed, use="pairwise.complete.obs") ) ) } # run 'correlateLengthSpeed' function with coasterData as a variable correlateLengthSpeed(coasterData) ##################################################################### # Task 1.5 # # - create function called 'compareWoodenSteal' tha accepts # # a dataframe as a variable # # - perform a T-test function to produce a p-level value # # - compare the results from the T-Test to find any significance# # at a level of 0.05 # # - use var.equal = FALSE to show unequal variance between the # # two columns # ##################################################################### # define 'compareWoodenSteel' function that accepts a dataframe variable compareWoodenSteel <- function(df) { # if statement to compare the results of the t-test over or under 0.05 if( # use in-built R function for t-test t.test( # define the columns that will be compared - speed and type speed ~ type, # input data for the t-test data = # omit any NA values within the data na.omit( data.frame( # define the type column with values steel and wooden type = rep(c("Steel", "Wooden")), # define the speed column isolating the columns from the dataframe speed = c(df[df$material_type=="Steel",]$speed,df[df$material_type=="Wooden",]$speed) ) ), # set var.equal to false as unequal variance var.equal=FALSE # return the p.value from the t-test )$p.value # compare the p.value to 0.05 <= 0.05) { # if p.value is less than or equal to 0.05 return 'Significant Difference' result <- "Significant Difference" } else { # if p.value is greaterthan 0.05 return 'No Significant Difference' result <- "No Significant Difference" } return(result) } # run 'compareWoodenSteal' function with coasterData as a variable compareWoodenSteel(coasterData)