Frederick Eze

Analysis of the Difference in Use of Cyclistic Bikes between Members and Casual Riders (2022 First Quarter) Frederick Eze Objective The objective of this business task is to identify how annual members and casual riders use cyclistic bikes differently. Data Source The dataset used for analysis consisted of data of each cyclistic bike ride, collected and stored on the cyclistic database. Data Cleaning and Manipulation Importing the Files This was done using R Studio and began with loading the tidyverse library and importing the csv files and assigning them to a variable name library(tidyverse) ctd_2022_01 <- read_csv("cyclistic_trip_data_2022/cyclistic_trip_data_2022_01/cyclistic_trip_data_2022_0 ctd_2022_02 <- read_csv("cyclistic_trip_data_2022/cyclistic_trip_data_2022_02/cyclistic_trip_data_2022_0 ctd_2022_03 <- read_csv("cyclistic_trip_data_2022/cyclistic_trip_data_2022_03/cyclistic_trip_data_2022_0 Joining the Tables Then joining the tables using the full_join function to join the three tables together join1<-full_join(ctd_2022_01,ctd_2022_02) ctd_2022<-full_join(join1,ctd_2022_03) Mutate The mutate function was used to add three columns; ride_length to calculate the duration of the ride, ride_date to seperate the start date from the time in the started_at column and day_of_week to record the day of the week that the ride started ctd_2022<-mutate(ctd_2022, ride_length=difftime(ended_at, started_at, units = "mins"), ride_date=ctd_202 ctd_2022<-mutate(ctd_2022, day_of_week=wday(ctd_2022$ride_date, label = TRUE)) head(ctd_2022) ## ## ## ## ## ## ## ## ## ## # A tibble: 6 x 16 ride_id rideable_type started_at ended_at 1 C2F7DD78E82EC875 electric_bike-:59:-:02:44 2 A6CF8980A652D272 electric_bike-:41:-:46:17 3 BD0F91DFF741C66D classic_bike-:53:-:58:01 4 CBB80ED- classic_bike-:18:-:33:00 5 DDC963BFDDA51EEA classic_bike-:31:-:37:12 6 A39C6F6CC0586C0B classic_bike-:48:-:51:31 # i 12 more variables: start_station_name , start_station_id , 1 ## # ## # ## # end_station_name , end_station_id , start_lat , start_lng , end_lat , end_lng , member_casual , ride_length , ride_date , day_of_week Filtering/Sorting and Making New Dataframes New data frames were created for the analysis of the number of rides and the average length of the rides and were named no_of_rides and avg_ride_length. no_of_rides<-ctd_2022 %>% select(ride_id, member_casual, rideable_type, day_of_week) %>% group_by(member_casual, rideable_type, day_of_week) %>% summarise(no_rides=n_distinct(ride_id)) avg_ride_length<-ctd_2022 %>% select(member_casual, rideable_type, ride_length, day_of_week) %>% group_by(day_of_week, member_casual, rideable_type) %>% summarise(avg_time=mean(ride_length)) head(no_of_rides) New Dataframe Output ## ## ## ## ## ## ## ## ## ## # A tibble: 6 x 4 # Groups: member_casual, rideable_type [1] member_casual rideable_type day_of_week no_rides 1 casual classic_bike Sun 9368 2 casual classic_bike Mon 8262 3 casual classic_bike Tue 5236 4 casual classic_bike Wed 7371 5 casual classic_bike Thu 5982 6 casual classic_bike Fri 4343 head(avg_ride_length) ## ## ## ## ## ## ## ## ## ## # A tibble: 6 x 4 # Groups: day_of_week, member_casual [3] day_of_week member_casual rideable_type avg_time 1 Sun casual classic_bike- Sun casual docked_bike- Sun casual electric_bike- Sun member classic_bike- Sun member electric_bike- Mon casual classic_bike- mins mins mins mins mins mins Analysis Total Number of Rides From the no_of_rides data frame, the total number of rides between annual members and casual riders was compared and it was found out that the total number of rides by annual members was significantly higher than the total number of rides by casual riders and is shown below ggplot(data=no_of_rides)+ geom_col(mapping = aes(x=member_casual, y=no_rides, fill=member_casual))+ labs(title = "Total Number of Rides of Member and Casual Riders", x="Member or Casual Rider", y="Numbe 2 Total Number of Rides of Member and Casual Riders Number of Rides 3e+05 member_casual 2e+05 casual member 1e+05 0e+00 casual member Member or Casual Rider Total Number of Rides on Each Bike Further analysis showed that annual members on average take more trips on the classic bikes followed by the electric bikes with no ride recorded on the docked bikes. Analysis also showed that on average, casual riders take the most trips on the electric bikes followed by the classic bikes with the docked bikes being used the least. ggplot(data=no_of_rides)+ geom_col(mapping = aes(x=member_casual, y=no_rides, fill=member_casual))+ facet_wrap(~rideable_type)+ labs(title = "Number of Rides on Each Bike", subtitle = "The total number of rides on each bike by mem 3 Number of Rides on Each Bike The total number of rides on each bike by member and casual riders classic_bike docked_bike electric_bike 200000 Number of Rides 150000 member_casual 100000 casual member 50000 0 casual member casual member casual member Member or Casual Number of Rides Per Day of Week Analysis showed that for annual members, the average number of rides on both the classic and electric bike during the week is higher than during the weekends while for casual riders, on the electric bike,the average number of rides remain steady during the week with a drop on Fridays, the classic bikes seem to be more active on the weekends and see a decline during the week while docked bikes seem to remain steady ggplot(data = no_of_rides)+ geom_col(mapping = aes(x=day_of_week, y=no_rides, fill=member_casual))+ facet_grid(rideable_type~member_casual)+ labs(title = "Number of Rides on Each Bike Per Day of Week", subtitle = "The total number of rides on theme(axis.text.x = element_text(angle = 45)) 4 Number of Rides on Each Bike Per Day of Week The total number of rides on each bike per day of week by member and casual riders casual member classic_bike - 0 30000 docked_bike Number of Rides 10000 - member_casual casual member 0 electric_bike - i Sa t Fr Th u W ed Tu e n on M Su t Sa i Fr Th u W ed Tu e on M Su n 0 Day of the Week Average Ride Length From the avg_ride_length dataframe,the average duration of rides between annual members and casual riders was compared and it was found out that the average duration of rides by casual riders was significantly longer than the average duration of rides by annual members and is shown below ggplot(data = avg_ride_length)+ geom_col(mapping = aes(x=member_casual, y=avg_time, fill=member_casual))+ labs(title = "Average Ride Length of Member and Casual Riders", x="Member or Casual", y="Average Time 5 Average Ride Length of Member and Casual Riders 1250 Average Time (mins) 1000 750 member_casual casual member 500 250 0 casual member Member or Casual Average Ride Length on Each Bike Further analysis showed that casual riders had a significantly longer ride duration on docked bikes than the other two, followed by the classic bikes with electric bikes having the shortest ride duration. It also showed that on average, annual members take longer trips on classic bikes than electric bikes but with no significant difference. ggplot(data = avg_ride_length)+ geom_col(mapping = aes(x=member_casual, y=avg_time, fill=member_casual))+ facet_wrap(~rideable_type)+ labs(title = "Average Ride Length on Each Bike", subtitle = "The average ride length on each bike by m 6 Average Ride Length on Each Bike The average ride length on each bike by member and casual riders classic_bike docked_bike electric_bike Average Time (mins) 750 member_casual 500 casual member 250 0 casual member casual member casual member Member or Casual Average Ride Length Per Day of Week for Each Bike Analysis showed that the average ride duration of casual riders on classic and electric bikes were relatively the same throughout the week with little fluctuation while on docked bikes, Thursday was higher than the rest of the days which might be influenced by factors unknown. The average ride duration of annual members on classic and electric bikes were also relatively the same throughout the week with little fluctuation. ggplot(data = avg_ride_length)+ geom_col(mapping = aes(x=day_of_week, y=avg_time, fill=member_casual))+ facet_grid(rideable_type~member_casual)+ labs(title = "Average Ride Length on Each Bike Per Day of Week", subtitle = "The average ride length o theme(axis.text.x = element_text(angle = 45)) 7 Average Ride Length on Each Bike Per Day of Week The average ride length on each bike per day of week by member and casual riders casual member classic_bike 150 100 0 150 docked_bike Average Time (mins) 50 100 50 member_casual casual member 0 electric_bike - t i Sa u Fr Th W ed Tu e n on M Su t i Sa u Fr Th W ed Tu e M Su n on 0 Day of the Week Conclusion • Annual members take more rides than casual riders • Annual members take more rides during the week while casual riders take more trips during the weekend • Annual members take the most trips on classic bikes while casual riders take the most trips on electric bikes • Only casual riders use docked bikes • Casual riders take longer trips than annual members • Casual riders use docked bikes for significantly longer trips 8