This case study, which is based on the analysis of a dataset from a fictional bike-sharing company called Cyclistic, is the capstone project of my Google Data Analytics Professional Certificate course.
In this project, I will be using the R programming language to conduct my analysis, complemented by the RStudio IDE for its robust ability to integrate effective statistical analysis tools with compelling data visualizations.
In order to answer the business questions of this case study, I will be following the data analysis process steps:
In each of these steps, the analysis will be broken down into:
The proposed scenario for this case study is that you are a junior data analyst working on the marketing analyst team at Cyclistic, a bike-share company in Chicago. The director of marketing believes the company’s future success depends on maximizing the number of annual memberships. Therefore, your team wants to understand how casual riders and annual members use Cyclistic bikes differently. From these insights, your team will design a new marketing strategy to convert casual riders into annual members. But first, Cyclistic executives must approve your recommendations, so they must be backed up with compelling data insights and professional data visualizations.
Three questions will guide the future marketing program of the company:
How do annual members and casual riders use Cyclistic bikes differently?
Why would casual riders buy Cyclistic annual memberships?
How can Cyclistic use digital media to influence casual riders to become members?
Moreno has assigned you the first question to answer: How do annual members and casual riders use Cyclistic bikes differently?
KEY TASKS:
Understand how casual riders and annual members use Cyclistic bikes differently.
Cyclistic Executive Team: Decision-makers who will approve the recommended marketing program.
Lily Moreno (Director of Marketing): Responsible for the development and implementation of marketing campaigns.
Cyclistic Marketing Analytics Team: Analysts who contribute to data collection, analysis, and reporting.
Cyclistic Users (Casual Riders and Annual Members): End-users whose behavior is under analysis for strategic marketing decisions.
DELIVERABLE:
Understand how casual riders and annual members use Cyclistic bikes differently, in order to design a new marketing strategy to drive future business success by maximizing the number of annual memberships, based on the analysis of the data.
In this second step, I will prepare the data by establishing where is it located, how is it organized, if there are any problems with the data or issues with bias or credibility. I will also be addressing the issues of licensing, privacy and security of the data, and verify the data’s integrity and credibility.
Finally, I will determine how can this data help answer the business questions.
I will use Cyclistic’s own historical trip data to analyze and identify trends. This is public data, which has been made available by Motivate International Inc. under this license. The respective dataset is available here.
KEY TASKS:
The data has been downloaded from this database, stored on my hard drive and the respective copies organized in this case study database folder, and made available on my RPubs and Kaggle profiles.
The data is in comma-delimeted (.CSV) format. It is composed of four files, the “Divvy_Trips_2019_Q2” with 12 columns and 1.048.575 rows; the “Divvy_Trips_2019_Q3” with also 12 columns and 1.048.575 rows; the “Divvy_Trips_2019_Q4” with 12 columns and 704.054 rows; and the “Divvy_Trips_2020_Q1” with 13 columns and 426.887 rows.
For this analysis I will work with the data from the second, third and fourth quarters of 2019 plus the 1st quarter of 2020, completing a 12-month period of data under this analysis.
The credibility of the data can be assessed using the ROCCC system, which shows that:
The data’s credibility and integrity is verifiable and, therefore, the data is adequate to provide reliable and comprehensive insights for this case study.
This data has limitations regarding data privacy issues, since it is not allowed to use riders’ personally identifiable information, such as gender and age. This means it won’t be possible, in the scope of this case study, to connect pass purchases to credit card numbers and determine the respective identifiers, i.e., if casual riders live in areas with Cyclistic services available or if they have purchased multiple single passes.
DELIVERABLE:
The data source is primary, provided by the Cyclistic company, and all data provided is available here.
In this section, I will start to process the data for analysis by gathering it, cleaning it and prepping it.
I will conduct tests to confirm the data’s integrity and I will also ensure that the data is clean and ready to analyze.
All of the process will be documented so that the results can be reviewed and shared.
KEY TASKS
Check the data for errors;
Choose the tools to use;
Transformation of the data so it can be worked with effectively;
Document the cleaning process.
The following R code chunks, presented in this section, will perform these tasks.
DELIVERABLE
The following R code chunks, presented in this section, will provide this deliverable.
CODE:
Installation and loading of the packages we’ll need:
## installation and loading of the packages
library(tidyverse) # helps wrangle data
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.4.4 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.0
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(conflicted) # Use the conflicted package to manage conflicts
library (readr) # readr function to load the csv files
# Set dplyr::filter and dplyr::lag as the default choices
conflict_prefer("filter", "dplyr")
## [conflicted] Will prefer dplyr::filter over any other package.
conflict_prefer("lag", "dplyr")
## [conflicted] Will prefer dplyr::lag over any other package.
Collecting and loading the datasets:
## uploading and collecting the Divvy datasets (csv files)
q2_2019 <- read_csv("Divvy_Trips_2019_Q2.csv")
## Warning: One or more parsing issues, call `problems()` on your data frame for details,
## e.g.:
## dat <- vroom(...)
## problems(dat)
## Rows: 1048575 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (5): 01 - Rental Details Rental ID, 03 - Rental Start Station Name, 02 ...
## dbl (5): 01 - Rental Details Bike ID, 01 - Rental Details Duration In Secon...
## dttm (2): 01 - Rental Details Local Start Time, 01 - Rental Details Local En...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
q3_2019 <- read_csv("Divvy_Trips_2019_Q3.csv")
## Warning: One or more parsing issues, call `problems()` on your data frame for details,
## e.g.:
## dat <- vroom(...)
## problems(dat)
## Rows: 1048575 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (5): trip_id, from_station_name, to_station_name, usertype, gender
## dbl (5): bikeid, tripduration, from_station_id, to_station_id, birthyear
## dttm (2): start_time, end_time
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
q4_2019 <- read_csv("Divvy_Trips_2019_Q4.csv")
## Warning: One or more parsing issues, call `problems()` on your data frame for details,
## e.g.:
## dat <- vroom(...)
## problems(dat)
## Rows: 704054 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (5): trip_id, from_station_name, to_station_name, usertype, gender
## dbl (5): bikeid, tripduration, from_station_id, to_station_id, birthyear
## dttm (2): start_time, end_time
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
q1_2020 <- read_csv("Divvy_Trips_2020_Q1.csv")
## Rows: 426887 Columns: 13
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (5): ride_id, rideable_type, start_station_name, end_station_name, memb...
## dbl (6): start_station_id, end_station_id, start_lat, start_lng, end_lat, e...
## dttm (2): started_at, ended_at
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Wrangling the data into a single file:
# Compare column names each of the files
# While the names don't have to be in the same order, they DO need to match perfectly before we can use a command to join them into one file
colnames(q2_2019)
## [1] "01 - Rental Details Rental ID"
## [2] "01 - Rental Details Local Start Time"
## [3] "01 - Rental Details Local End Time"
## [4] "01 - Rental Details Bike ID"
## [5] "01 - Rental Details Duration In Seconds Uncapped"
## [6] "03 - Rental Start Station ID"
## [7] "03 - Rental Start Station Name"
## [8] "02 - Rental End Station ID"
## [9] "02 - Rental End Station Name"
## [10] "User Type"
## [11] "Member Gender"
## [12] "05 - Member Details Member Birthday Year"
colnames(q3_2019)
## [1] "trip_id" "start_time" "end_time"
## [4] "bikeid" "tripduration" "from_station_id"
## [7] "from_station_name" "to_station_id" "to_station_name"
## [10] "usertype" "gender" "birthyear"
colnames(q4_2019)
## [1] "trip_id" "start_time" "end_time"
## [4] "bikeid" "tripduration" "from_station_id"
## [7] "from_station_name" "to_station_id" "to_station_name"
## [10] "usertype" "gender" "birthyear"
colnames(q1_2020)
## [1] "ride_id" "rideable_type" "started_at"
## [4] "ended_at" "start_station_name" "start_station_id"
## [7] "end_station_name" "end_station_id" "start_lat"
## [10] "start_lng" "end_lat" "end_lng"
## [13] "member_casual"
Finding: Column names between the files are not the same nor consistent. They need to match perfectly in order to be joined into a single file. It is, therefore, necessary to rename the divergent columns for congruency, which will be accomplished with the following code chunks:
# Rename columns of q2_2019, q3_2019 and q4_2019 to make them consistent with q1_2020
(q2_2019 <- rename(q2_2019
,ride_id = '01 - Rental Details Rental ID'
,rideable_type = '01 - Rental Details Bike ID'
,started_at = '01 - Rental Details Local Start Time'
,ended_at = '01 - Rental Details Local End Time'
,start_station_name = '03 - Rental Start Station Name'
,start_station_id = '03 - Rental Start Station ID'
,end_station_name = '02 - Rental End Station Name'
,end_station_id = '02 - Rental End Station ID'
,member_casual = 'User Type'))
## # A tibble: 1,048,575 × 12
## ride_id started_at ended_at rideable_type
## <chr> <dttm> <dttm> <dbl>
## 1 "22178529" 2019-04-01 00:02:22 2019-04-01 00:09:48 6251
## 2 "22178530,2019-04-01 0… NA NA NA
## 3 "22178531" 2019-04-01 00:11:07 2019-04-01 00:15:19 5649
## 4 "22178532" 2019-04-01 00:13:01 2019-04-01 00:18:58 4151
## 5 "22178533,2019-04-01 0… NA NA NA
## 6 "22178534" 2019-04-01 00:19:39 2019-04-01 00:23:56 3123
## 7 "22178535" 2019-04-01 00:26:33 2019-04-01 00:35:41 6418
## 8 "22178536" 2019-04-01 00:29:48 2019-04-01 00:36:11 4513
## 9 "22178537,2019-04-01 0… NA NA NA
## 10 "22178538,2019-04-01 0… NA NA NA
## # ℹ 1,048,565 more rows
## # ℹ 8 more variables: `01 - Rental Details Duration In Seconds Uncapped` <dbl>,
## # start_station_id <dbl>, start_station_name <chr>, end_station_id <dbl>,
## # end_station_name <chr>, member_casual <chr>, `Member Gender` <chr>,
## # `05 - Member Details Member Birthday Year` <dbl>
(q3_2019 <- rename(q3_2019
,ride_id = trip_id
,rideable_type = bikeid
,started_at = start_time
,ended_at = end_time
,start_station_name = from_station_name
,start_station_id = from_station_id
,end_station_name = to_station_name
,end_station_id = to_station_id
,member_casual = usertype))
## # A tibble: 1,048,575 × 12
## ride_id started_at ended_at rideable_type tripduration
## <chr> <dttm> <dttm> <dbl> <dbl>
## 1 "23479388… NA NA NA NA
## 2 "23479389… NA NA NA NA
## 3 "23479390… NA NA NA NA
## 4 "23479391… NA NA NA NA
## 5 "23479392… NA NA NA NA
## 6 "23479393" 2019-07-01 00:02:21 2019-07-01 00:07:31 4941 310
## 7 "23479394… NA NA NA NA
## 8 "23479395… NA NA NA NA
## 9 "23479396… NA NA NA NA
## 10 "23479397… NA NA NA NA
## # ℹ 1,048,565 more rows
## # ℹ 7 more variables: start_station_id <dbl>, start_station_name <chr>,
## # end_station_id <dbl>, end_station_name <chr>, member_casual <chr>,
## # gender <chr>, birthyear <dbl>
(q4_2019 <- rename(q4_2019
,ride_id = trip_id
,rideable_type = bikeid
,started_at = start_time
,ended_at = end_time
,start_station_name = from_station_name
,start_station_id = from_station_id
,end_station_name = to_station_name
,end_station_id = to_station_id
,member_casual = usertype))
## # A tibble: 704,054 × 12
## ride_id started_at ended_at rideable_type tripduration
## <chr> <dttm> <dttm> <dbl> <dbl>
## 1 "25223640" 2019-10-01 00:01:39 2019-10-01 00:17:20 2215 940
## 2 "25223641" 2019-10-01 00:02:16 2019-10-01 00:06:34 6328 258
## 3 "25223642" 2019-10-01 00:04:32 2019-10-01 00:18:43 3003 850
## 4 "25223643… NA NA NA NA
## 5 "25223644… NA NA NA NA
## 6 "25223645" 2019-10-01 00:04:38 2019-10-01 00:10:51 1891 373
## 7 "25223646… NA NA NA NA
## 8 "25223647… NA NA NA NA
## 9 "25223648… NA NA NA NA
## 10 "25223649… NA NA NA NA
## # ℹ 704,044 more rows
## # ℹ 7 more variables: start_station_id <dbl>, start_station_name <chr>,
## # end_station_id <dbl>, end_station_name <chr>, member_casual <chr>,
## # gender <chr>, birthyear <dbl>
Let’s inspect the dataframes and look for incongruencies
str(q2_2019)
## spc_tbl_ [1,048,575 × 12] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
## $ ride_id : chr [1:1048575] "22178529" "22178530,2019-04-01 00:03:02,2019-04-01 00:20:30,6226,\"1,048.0\",317,Wood St & Taylor St,59,Wabash Ave & Roose"| __truncated__ "22178531" "22178532" ...
## $ started_at : POSIXct[1:1048575], format: "2019-04-01 00:02:22" NA ...
## $ ended_at : POSIXct[1:1048575], format: "2019-04-01 00:09:48" NA ...
## $ rideable_type : num [1:1048575] 6251 NA 5649 4151 NA ...
## $ 01 - Rental Details Duration In Seconds Uncapped: num [1:1048575] 446 NA 252 357 NA 257 548 383 NA NA ...
## $ start_station_id : num [1:1048575] 81 NA 283 26 NA 420 503 260 NA NA ...
## $ start_station_name : chr [1:1048575] "Daley Center Plaza" NA "LaSalle St & Jackson Blvd" "McClurg Ct & Illinois St" ...
## $ end_station_id : num [1:1048575] 56 NA 174 133 NA 426 500 499 NA NA ...
## $ end_station_name : chr [1:1048575] "Desplaines St & Kinzie St" NA "Canal St & Madison St" "Kingsbury St & Kinzie St" ...
## $ member_casual : chr [1:1048575] "Subscriber" NA "Subscriber" "Subscriber" ...
## $ Member Gender : chr [1:1048575] "Male" NA "Male" "Male" ...
## $ 05 - Member Details Member Birthday Year : num [1:1048575] 1975 NA 1990 1993 NA ...
## - attr(*, "spec")=
## .. cols(
## .. `01 - Rental Details Rental ID` = col_character(),
## .. `01 - Rental Details Local Start Time` = col_datetime(format = ""),
## .. `01 - Rental Details Local End Time` = col_datetime(format = ""),
## .. `01 - Rental Details Bike ID` = col_double(),
## .. `01 - Rental Details Duration In Seconds Uncapped` = col_double(),
## .. `03 - Rental Start Station ID` = col_double(),
## .. `03 - Rental Start Station Name` = col_character(),
## .. `02 - Rental End Station ID` = col_double(),
## .. `02 - Rental End Station Name` = col_character(),
## .. `User Type` = col_character(),
## .. `Member Gender` = col_character(),
## .. `05 - Member Details Member Birthday Year` = col_double()
## .. )
## - attr(*, "problems")=<externalptr>
str(q3_2019)
## spc_tbl_ [1,048,575 × 12] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
## $ ride_id : chr [1:1048575] "23479388,2019-07-01 00:00:27,2019-07-01 00:20:41,3591,\"1,214.0\",117,Wilton Ave & Belmont Ave,497,Kimball Ave "| __truncated__ "23479389,2019-07-01 00:01:16,2019-07-01 00:18:44,5353,\"1,048.0\",381,Western Ave & Monroe St,203,Western Ave &"| __truncated__ "23479390,2019-07-01 00:01:48,2019-07-01 00:27:42,6180,\"1,554.0\",313,Lakeview Ave & Fullerton Pkwy,144,Larrabe"| __truncated__ "23479391,2019-07-01 00:02:07,2019-07-01 00:27:10,5540,\"1,503.0\",313,Lakeview Ave & Fullerton Pkwy,144,Larrabe"| __truncated__ ...
## $ started_at : POSIXct[1:1048575], format: NA NA ...
## $ ended_at : POSIXct[1:1048575], format: NA NA ...
## $ rideable_type : num [1:1048575] NA NA NA NA NA ...
## $ tripduration : num [1:1048575] NA NA NA NA NA 310 NA NA NA NA ...
## $ start_station_id : num [1:1048575] NA NA NA NA NA 300 NA NA NA NA ...
## $ start_station_name: chr [1:1048575] NA NA NA NA ...
## $ end_station_id : num [1:1048575] NA NA NA NA NA 232 NA NA NA NA ...
## $ end_station_name : chr [1:1048575] NA NA NA NA ...
## $ member_casual : chr [1:1048575] NA NA NA NA ...
## $ gender : chr [1:1048575] NA NA NA NA ...
## $ birthyear : num [1:1048575] NA NA NA NA NA 1990 NA NA NA NA ...
## - attr(*, "spec")=
## .. cols(
## .. trip_id = col_character(),
## .. start_time = col_datetime(format = ""),
## .. end_time = col_datetime(format = ""),
## .. bikeid = col_double(),
## .. tripduration = col_double(),
## .. from_station_id = col_double(),
## .. from_station_name = col_character(),
## .. to_station_id = col_double(),
## .. to_station_name = col_character(),
## .. usertype = col_character(),
## .. gender = col_character(),
## .. birthyear = col_double()
## .. )
## - attr(*, "problems")=<externalptr>
str(q4_2019)
## spc_tbl_ [704,054 × 12] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
## $ ride_id : chr [1:704054] "25223640" "25223641" "25223642" "25223643,2019-10-01 00:04:32,2019-10-01 00:43:43,3275,\"2,350.0\",313,Lakeview Ave & Fullerton Pkwy,290,Kedzie "| __truncated__ ...
## $ started_at : POSIXct[1:704054], format: "2019-10-01 00:01:39" "2019-10-01 00:02:16" ...
## $ ended_at : POSIXct[1:704054], format: "2019-10-01 00:17:20" "2019-10-01 00:06:34" ...
## $ rideable_type : num [1:704054] 2215 6328 3003 NA NA ...
## $ tripduration : num [1:704054] 940 258 850 NA NA 373 NA NA NA NA ...
## $ start_station_id : num [1:704054] 20 19 84 NA NA 156 NA NA NA NA ...
## $ start_station_name: chr [1:704054] "Sheffield Ave & Kingsbury St" "Throop (Loomis) St & Taylor St" "Milwaukee Ave & Grand Ave" NA ...
## $ end_station_id : num [1:704054] 309 241 199 NA NA 226 NA NA NA NA ...
## $ end_station_name : chr [1:704054] "Leavitt St & Armitage Ave" "Morgan St & Polk St" "Wabash Ave & Grand Ave" NA ...
## $ member_casual : chr [1:704054] "Subscriber" "Subscriber" "Subscriber" NA ...
## $ gender : chr [1:704054] "Male" "Male" "Female" NA ...
## $ birthyear : num [1:704054] 1987 1998 1991 NA NA ...
## - attr(*, "spec")=
## .. cols(
## .. trip_id = col_character(),
## .. start_time = col_datetime(format = ""),
## .. end_time = col_datetime(format = ""),
## .. bikeid = col_double(),
## .. tripduration = col_double(),
## .. from_station_id = col_double(),
## .. from_station_name = col_character(),
## .. to_station_id = col_double(),
## .. to_station_name = col_character(),
## .. usertype = col_character(),
## .. gender = col_character(),
## .. birthyear = col_double()
## .. )
## - attr(*, "problems")=<externalptr>
str(q1_2020)
## spc_tbl_ [426,887 × 13] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
## $ ride_id : chr [1:426887] "EACB19130B0CDA4A" "8FED874C809DC021" "789F3C21E472CA96" "C9A388DAC6ABF313" ...
## $ rideable_type : chr [1:426887] "docked_bike" "docked_bike" "docked_bike" "docked_bike" ...
## $ started_at : POSIXct[1:426887], format: "2020-01-21 20:06:59" "2020-01-30 14:22:39" ...
## $ ended_at : POSIXct[1:426887], format: "2020-01-21 20:14:30" "2020-01-30 14:26:22" ...
## $ start_station_name: chr [1:426887] "Western Ave & Leland Ave" "Clark St & Montrose Ave" "Broadway & Belmont Ave" "Clark St & Randolph St" ...
## $ start_station_id : num [1:426887] 239 234 296 51 66 212 96 96 212 38 ...
## $ end_station_name : chr [1:426887] "Clark St & Leland Ave" "Southport Ave & Irving Park Rd" "Wilton Ave & Belmont Ave" "Fairbanks Ct & Grand Ave" ...
## $ end_station_id : num [1:426887] 326 318 117 24 212 96 212 212 96 100 ...
## $ start_lat : num [1:426887] 42 42 41.9 41.9 41.9 ...
## $ start_lng : num [1:426887] -87.7 -87.7 -87.6 -87.6 -87.6 ...
## $ end_lat : num [1:426887] 42 42 41.9 41.9 41.9 ...
## $ end_lng : num [1:426887] -87.7 -87.7 -87.7 -87.6 -87.6 ...
## $ member_casual : chr [1:426887] "member" "member" "member" "member" ...
## - attr(*, "spec")=
## .. cols(
## .. ride_id = col_character(),
## .. rideable_type = col_character(),
## .. started_at = col_datetime(format = ""),
## .. ended_at = col_datetime(format = ""),
## .. start_station_name = col_character(),
## .. start_station_id = col_double(),
## .. end_station_name = col_character(),
## .. end_station_id = col_double(),
## .. start_lat = col_double(),
## .. start_lng = col_double(),
## .. end_lat = col_double(),
## .. end_lng = col_double(),
## .. member_casual = col_character()
## .. )
## - attr(*, "problems")=<externalptr>
Finding: The column values ride_id and rideable_type need to be converted to character so that they can stack correctly, which will be accomplished with the following code chunk:
# Converting ride_id and rideable_type to character
q2_2019 <- mutate(q2_2019, ride_id = as.character(ride_id)
,rideable_type = as.character(rideable_type))
q3_2019 <- mutate(q3_2019, ride_id = as.character(ride_id)
,rideable_type = as.character(rideable_type))
q4_2019 <- mutate(q4_2019, ride_id = as.character(ride_id)
,rideable_type = as.character(rideable_type))
Now, we are going to stack all individual quarter’s data frames into one big data frame:
all_rides <- bind_rows(q2_2019, q3_2019, q4_2019, q1_2020)
Finding: Latitude values, longitude values, birthyear, and gender fields need to be removed from the dataset, since they were dropped in the beginning of 2020. The following code chunk performs this action:
# Remove extra columns that were dropped in the most recent data of 2020
all_rides <- all_rides %>%
select(-c('01 - Rental Details Duration In Seconds Uncapped','05 - Member Details Member Birthday Year', 'Member Gender','05 - Member Details Member Birthday Year', tripduration, start_lat, start_lng, end_lat, end_lng, birthyear, gender))
Now, let’s start the process of cleaning up and adding data to prepare for analysis. - Let´s start by inspecting the new table we created:
colnames(all_rides) #List of column names
## [1] "ride_id" "started_at" "ended_at"
## [4] "rideable_type" "start_station_id" "start_station_name"
## [7] "end_station_id" "end_station_name" "member_casual"
nrow(all_rides) #How many rows are in data frame?
## [1] 3228091
dim(all_rides) #Dimensions of the data frame?
## [1] 3228091 9
head(all_rides) #See the first 6 rows of data frame.
## # A tibble: 6 × 9
## ride_id started_at ended_at rideable_type start_station_id
## <chr> <dttm> <dttm> <chr> <dbl>
## 1 "22178… 2019-04-01 00:02:22 2019-04-01 00:09:48 6251 81
## 2 "22178… NA NA <NA> NA
## 3 "22178… 2019-04-01 00:11:07 2019-04-01 00:15:19 5649 283
## 4 "22178… 2019-04-01 00:13:01 2019-04-01 00:18:58 4151 26
## 5 "22178… NA NA <NA> NA
## 6 "22178… 2019-04-01 00:19:39 2019-04-01 00:23:56 3123 420
## # ℹ 4 more variables: start_station_name <chr>, end_station_id <dbl>,
## # end_station_name <chr>, member_casual <chr>
str(all_rides) #See list of columns and data types
## tibble [3,228,091 × 9] (S3: tbl_df/tbl/data.frame)
## $ ride_id : chr [1:3228091] "22178529" "22178530,2019-04-01 00:03:02,2019-04-01 00:20:30,6226,\"1,048.0\",317,Wood St & Taylor St,59,Wabash Ave & Roose"| __truncated__ "22178531" "22178532" ...
## $ started_at : POSIXct[1:3228091], format: "2019-04-01 00:02:22" NA ...
## $ ended_at : POSIXct[1:3228091], format: "2019-04-01 00:09:48" NA ...
## $ rideable_type : chr [1:3228091] "6251" NA "5649" "4151" ...
## $ start_station_id : num [1:3228091] 81 NA 283 26 NA 420 503 260 NA NA ...
## $ start_station_name: chr [1:3228091] "Daley Center Plaza" NA "LaSalle St & Jackson Blvd" "McClurg Ct & Illinois St" ...
## $ end_station_id : num [1:3228091] 56 NA 174 133 NA 426 500 499 NA NA ...
## $ end_station_name : chr [1:3228091] "Desplaines St & Kinzie St" NA "Canal St & Madison St" "Kingsbury St & Kinzie St" ...
## $ member_casual : chr [1:3228091] "Subscriber" NA "Subscriber" "Subscriber" ...
summary(all_rides) #Statistical summary of data. Mainly for numerics
## ride_id started_at
## Length:3228091 Min. :2019-04-01 00:02:22.00
## Class :character 1st Qu.:2019-06-16 00:53:27.00
## Mode :character Median :2019-08-11 23:15:22.00
## Mean :2019-09-08 05:48:12.92
## 3rd Qu.:2019-11-27 08:57:45.00
## Max. :2020-03-31 23:51:34.00
## NA's :1004094
## ended_at rideable_type start_station_id
## Min. :2019-04-01 00:09:48.00 Length:3228091 Min. : 1.0
## 1st Qu.:2019-06-16 01:05:40.00 Class :character 1st Qu.: 77.0
## Median :2019-08-11 23:24:24.00 Mode :character Median :174.0
## Mean :2019-09-08 05:59:27.44 Mean :203.5
## 3rd Qu.:2019-11-27 09:05:32.00 3rd Qu.:290.0
## Max. :2020-05-19 20:10:34.00 Max. :675.0
## NA's :1004094 NA's :1004094
## start_station_name end_station_id end_station_name member_casual
## Length:3228091 Min. : 1.0 Length:3228091 Length:3228091
## Class :character 1st Qu.: 77.0 Class :character Class :character
## Mode :character Median :174.0 Mode :character Mode :character
## Mean :203.4
## 3rd Qu.:290.0
## Max. :675.0
## NA's :1004095
Finding: From this quick inspection of the data, we can verify there are at least 4 problems that need to be fixed:
1) In the member_casual column, there are two names for members (“member” and “subscriber”) plus two names for casual riders (“customer” and “casual”). This needs to be consolidate into just two labels.
Let’s start by assessing how many observations fall under each usertype:
table(all_rides$member_casual)
##
## casual Customer member Subscriber
## 48480 189130 378407 1607980
Now, let’s reassign to the desired values (we will go with the current 2020 labels):
all_rides <- all_rides %>%
mutate(member_casual = recode(member_casual
,"Subscriber" = "member"
,"Customer" = "casual"))
Finaly, Let’s check to make sure the proper number of observations were reassigned:
table(all_rides$member_casual)
##
## casual member
## 237610 1986387
There, now we have just the two values to designate the type of user of the Cyclistic bike sharing service, members and casual riders.
2) The data can only be aggregated at the ride-level, which is too granular. It’s necessary to add some additional columns of data – such as day, month, year – that provide additional opportunities to aggregate the data.
# Add columns that list the date, month, day, and year of each ride
all_rides$date <- as.Date(all_rides$started_at) #The default format is yyyy-mm-dd
all_rides$month <- format(as.Date(all_rides$date), "%m")
all_rides$day <- format(as.Date(all_rides$date), "%d")
all_rides$year <- format(as.Date(all_rides$date), "%Y")
all_rides$day_of_week <- format(as.Date(all_rides$date), "%A")
3) We will also want to add a calculated field for length_of_ride since the 2020Q1 data didn’t have the tripduration column. I will add ride_length to the entire dataframe for consistency.
# Add a "ride_length" calculation to all_rides (in seconds)
all_rides$ride_length <- difftime(all_rides$ended_at,all_rides$started_at)
Let’s have a look at the structure of the columns to verify the creation of ride_length
str(all_rides)
## tibble [3,228,091 × 15] (S3: tbl_df/tbl/data.frame)
## $ ride_id : chr [1:3228091] "22178529" "22178530,2019-04-01 00:03:02,2019-04-01 00:20:30,6226,\"1,048.0\",317,Wood St & Taylor St,59,Wabash Ave & Roose"| __truncated__ "22178531" "22178532" ...
## $ started_at : POSIXct[1:3228091], format: "2019-04-01 00:02:22" NA ...
## $ ended_at : POSIXct[1:3228091], format: "2019-04-01 00:09:48" NA ...
## $ rideable_type : chr [1:3228091] "6251" NA "5649" "4151" ...
## $ start_station_id : num [1:3228091] 81 NA 283 26 NA 420 503 260 NA NA ...
## $ start_station_name: chr [1:3228091] "Daley Center Plaza" NA "LaSalle St & Jackson Blvd" "McClurg Ct & Illinois St" ...
## $ end_station_id : num [1:3228091] 56 NA 174 133 NA 426 500 499 NA NA ...
## $ end_station_name : chr [1:3228091] "Desplaines St & Kinzie St" NA "Canal St & Madison St" "Kingsbury St & Kinzie St" ...
## $ member_casual : chr [1:3228091] "member" NA "member" "member" ...
## $ date : Date[1:3228091], format: "2019-04-01" NA ...
## $ month : chr [1:3228091] "04" NA "04" "04" ...
## $ day : chr [1:3228091] "01" NA "01" "01" ...
## $ year : chr [1:3228091] "2019" NA "2019" "2019" ...
## $ day_of_week : chr [1:3228091] "segunda-feira" NA "segunda-feira" "segunda-feira" ...
## $ ride_length : 'difftime' num [1:3228091] 446 NA 252 357 ...
## ..- attr(*, "units")= chr "secs"
In order for us to run calculations on it, we need to convert ride_length from factor to numeric Let’s to that now:
is.factor(all_rides$ride_length)
## [1] FALSE
all_rides$ride_length <- as.numeric(as.character(all_rides$ride_length))
is.numeric(all_rides$ride_length)
## [1] TRUE
Let’s inspect the structure of the columns to make sure that ride_length is now numeric
str(all_rides)
## tibble [3,228,091 × 15] (S3: tbl_df/tbl/data.frame)
## $ ride_id : chr [1:3228091] "22178529" "22178530,2019-04-01 00:03:02,2019-04-01 00:20:30,6226,\"1,048.0\",317,Wood St & Taylor St,59,Wabash Ave & Roose"| __truncated__ "22178531" "22178532" ...
## $ started_at : POSIXct[1:3228091], format: "2019-04-01 00:02:22" NA ...
## $ ended_at : POSIXct[1:3228091], format: "2019-04-01 00:09:48" NA ...
## $ rideable_type : chr [1:3228091] "6251" NA "5649" "4151" ...
## $ start_station_id : num [1:3228091] 81 NA 283 26 NA 420 503 260 NA NA ...
## $ start_station_name: chr [1:3228091] "Daley Center Plaza" NA "LaSalle St & Jackson Blvd" "McClurg Ct & Illinois St" ...
## $ end_station_id : num [1:3228091] 56 NA 174 133 NA 426 500 499 NA NA ...
## $ end_station_name : chr [1:3228091] "Desplaines St & Kinzie St" NA "Canal St & Madison St" "Kingsbury St & Kinzie St" ...
## $ member_casual : chr [1:3228091] "member" NA "member" "member" ...
## $ date : Date[1:3228091], format: "2019-04-01" NA ...
## $ month : chr [1:3228091] "04" NA "04" "04" ...
## $ day : chr [1:3228091] "01" NA "01" "01" ...
## $ year : chr [1:3228091] "2019" NA "2019" "2019" ...
## $ day_of_week : chr [1:3228091] "segunda-feira" NA "segunda-feira" "segunda-feira" ...
## $ ride_length : num [1:3228091] 446 NA 252 357 NA 257 548 383 NA NA ...
It is.
For the sake of clarity of the analysis and presentation of the results we will turn ride_length from seconds to minutes:
all_rides$ride_length <- all_rides$ride_length/60
Let’s have another quick inspection to verify the structure of the columns and confirm that ride_length now shows in minutes not seconds.
str(all_rides)
## tibble [3,228,091 × 15] (S3: tbl_df/tbl/data.frame)
## $ ride_id : chr [1:3228091] "22178529" "22178530,2019-04-01 00:03:02,2019-04-01 00:20:30,6226,\"1,048.0\",317,Wood St & Taylor St,59,Wabash Ave & Roose"| __truncated__ "22178531" "22178532" ...
## $ started_at : POSIXct[1:3228091], format: "2019-04-01 00:02:22" NA ...
## $ ended_at : POSIXct[1:3228091], format: "2019-04-01 00:09:48" NA ...
## $ rideable_type : chr [1:3228091] "6251" NA "5649" "4151" ...
## $ start_station_id : num [1:3228091] 81 NA 283 26 NA 420 503 260 NA NA ...
## $ start_station_name: chr [1:3228091] "Daley Center Plaza" NA "LaSalle St & Jackson Blvd" "McClurg Ct & Illinois St" ...
## $ end_station_id : num [1:3228091] 56 NA 174 133 NA 426 500 499 NA NA ...
## $ end_station_name : chr [1:3228091] "Desplaines St & Kinzie St" NA "Canal St & Madison St" "Kingsbury St & Kinzie St" ...
## $ member_casual : chr [1:3228091] "member" NA "member" "member" ...
## $ date : Date[1:3228091], format: "2019-04-01" NA ...
## $ month : chr [1:3228091] "04" NA "04" "04" ...
## $ day : chr [1:3228091] "01" NA "01" "01" ...
## $ year : chr [1:3228091] "2019" NA "2019" "2019" ...
## $ day_of_week : chr [1:3228091] "segunda-feira" NA "segunda-feira" "segunda-feira" ...
## $ ride_length : num [1:3228091] 7.43 NA 4.2 5.95 NA ...
And now it does.
4) Another issue with the data is that there are some rides where tripduration shows up as negative, including several hundred rides where bikes were taken out of circulation for Quality Control reasons. These rides will have to be deleted.
Let’s do that right now:
all_rides_v2 <- all_rides[!(all_rides$start_station_name == "HQ QR" | all_rides$ride_length<0),]
For the sake of clarity of the analysis and cleaning of the data, I will also remove the nulls from the dataset. We could use drop_na() %>% in each of the relevant commands, but in this case I opted to clean the dataset right off the bat.
all_rides_v2 <- na.omit(all_rides_v2)
Let’s inspect the structure of our new dataframe to confirm that there are no null values in any of the columns:
summary(all_rides_v2)
## ride_id started_at
## Length:2220220 Min. :2019-04-01 00:02:22.00
## Class :character 1st Qu.:2019-06-15 18:28:41.75
## Mode :character Median :2019-08-11 17:39:55.00
## Mean :2019-09-07 22:21:36.69
## 3rd Qu.:2019-11-26 13:49:17.50
## Max. :2020-03-31 23:51:34.00
## ended_at rideable_type start_station_id
## Min. :2019-04-01 00:09:48.00 Length:2220220 Min. : 1.0
## 1st Qu.:2019-06-15 18:39:00.25 Class :character 1st Qu.: 77.0
## Median :2019-08-11 17:49:30.50 Mode :character Median :174.0
## Mean :2019-09-07 22:32:52.35 Mean :202.7
## 3rd Qu.:2019-11-26 13:55:34.00 3rd Qu.:289.0
## Max. :2020-05-19 20:10:34.00 Max. :673.0
## start_station_name end_station_id end_station_name member_casual
## Length:2220220 Min. : 1.0 Length:2220220 Length:2220220
## Class :character 1st Qu.: 77.0 Class :character Class :character
## Mode :character Median :174.0 Mode :character Mode :character
## Mean :202.6
## 3rd Qu.:289.0
## Max. :675.0
## date month day year
## Min. :2019-04-01 Length:2220220 Length:2220220 Length:2220220
## 1st Qu.:2019-06-15 Class :character Class :character Class :character
## Median :2019-08-11 Mode :character Mode :character Mode :character
## Mean :2019-09-07
## 3rd Qu.:2019-11-26
## Max. :2020-03-31
## day_of_week ride_length
## Length:2220220 Min. : 0.02
## Class :character 1st Qu.: 5.50
## Mode :character Median : 8.42
## Mean : 11.26
## 3rd Qu.: 12.10
## Max. :156450.40
And now we have no null values littering our dataset.
Now that the data is stored appropriately and has been prepared for analysis, it’s time to start putting it to work.
In this section, I will organize and properly format the data in order to perform analysis on it, look for significant trends, relationships and generate insights from it, to help answer the business task.
KEY TASKS
Aggregate the data so it’s useful and accessible.
Organize and format the data.
Perform calculations.
Identify trends and relationships.
The following R code chunks, presented in this section, will perform these tasks.
DELIVERABLE
Let’s start by conducting our descriptive analysis.
First, we’ll have a look at the ride lengths to see if the data has anything interesting to tell us, in this aspect.
CODE
# Descriptive analysis on ride_length (all figures in minutes)
mean(all_rides_v2$ride_length) #straight average (total ride length / rides)
## [1] 11.26118
median(all_rides_v2$ride_length) #midpoint in the ascending array of ride lengths
## [1] 8.416667
max(all_rides_v2$ride_length) #longest ride
## [1] 156450.4
min(all_rides_v2$ride_length) #shortest ride
## [1] 0.01666667
# Or using summary() on the specific attribute to condense the above 4 line in 1
summary(all_rides_v2$ride_length)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.02 5.50 8.42 11.26 12.10 156450.40
Finding: From these quick calculations, it stands out, for instances, that the average ride time for all types of riders is of 11 minutes and 26 seconds.
Now, let’s find out with the same metrics, if there is a big discrepancy or not in ride times between members vs casual users.
# Compare mean, median, max and min of ride_legth between members vs casual users
aggregate(all_rides_v2$ride_length ~ all_rides_v2$member_casual, FUN = mean)
## all_rides_v2$member_casual all_rides_v2$ride_length
## 1 casual 28.56247
## 2 member 9.22444
aggregate(all_rides_v2$ride_length ~ all_rides_v2$member_casual, FUN = median)
## all_rides_v2$member_casual all_rides_v2$ride_length
## 1 casual 11.81667
## 2 member 8.05000
aggregate(all_rides_v2$ride_length ~ all_rides_v2$member_casual, FUN = max)
## all_rides_v2$member_casual all_rides_v2$ride_length
## 1 casual 156450.40
## 2 member 93793.52
aggregate(all_rides_v2$ride_length ~ all_rides_v2$member_casual, FUN = min)
## all_rides_v2$member_casual all_rides_v2$ride_length
## 1 casual 0.03333333
## 2 member 0.01666667
Finding: There is a big difference between the ride times of members vs casual users, where casual user rides are much longer, in terms of ride time, than member user rides.
Now, let’s find out if this varies in terms of the day of the week or if this difference in ride times between members and casual users holds regardless the day of the week.
# Let’s assess the average ride time by each day for members vs casual users
aggregate(all_rides_v2$ride_length ~ all_rides_v2$member_casual + all_rides_v2$day_of_week, FUN = mean)
## all_rides_v2$member_casual all_rides_v2$day_of_week all_rides_v2$ride_length
## 1 casual domingo 36.457537
## 2 member domingo 10.011278
## 3 casual quarta-feira 24.700810
## 4 member quarta-feira 9.036617
## 5 casual quinta-feira 30.820628
## 6 member quinta-feira 8.999737
## 7 casual sábado 26.040054
## 8 member sábado 9.714433
## 9 casual segunda-feira 23.014277
## 10 member segunda-feira 9.270657
## 11 casual sexta-feira 28.151452
## 12 member sexta-feira 9.124590
## 13 casual terça-feira 25.641588
## 14 member terça-feira 9.025934
Due to my geolocation, the days of the week appear in portuguese.Here is a quick translation:
Finding: As we can confirm this trend holds across all days of the week, casual user rides are in average always much longer in terms of ride time, than member user rides.
So, we have confirmed that on average rides taken by casual users of Cyclistic are much longer than those taken by members.
Now, let’s explore the number of rides – who takes more rides, casual users or members?
Let’s find out.
all_rides_v2 %>%
group_by(member_casual) %>% # groups by type of user
summarise(number_of_rides = n()) %>% # calculates the number of rides
arrange(member_casual) # sorts it by type of user
## # A tibble: 2 × 2
## member_casual number_of_rides
## <chr> <int>
## 1 casual 233840
## 2 member 1986380
Finding: As we can confirm, in terms of the number of rides made, members make significantly more rides in total than casual riders—precisely 89.47% more rides.
Now, let’s verify if this is a trend that holds throughout the days of the week, or if in some days there are more registered rides from casual users than from members, for example on weekends.
Let’s check it out.
all_rides_v2 %>%
mutate(weekday = wday(started_at, label = TRUE)) %>% # creates the weekday field
group_by(member_casual, weekday) %>% #groups by usertype and weekday
summarise(number_of_rides = n()) %>% # calculates the number of rides
arrange(member_casual, weekday) # sorts it
## `summarise()` has grouped output by 'member_casual'. You can override using the
## `.groups` argument.
## # A tibble: 14 × 3
## # Groups: member_casual [2]
## member_casual weekday number_of_rides
## <chr> <ord> <int>
## 1 casual dom 49154
## 2 casual seg 25719
## 3 casual ter 26391
## 4 casual qua 27666
## 5 casual qui 28303
## 6 casual sex 32142
## 7 casual sáb 44465
## 8 member dom 172335
## 9 member seg 323335
## 10 member ter 350426
## 11 member qua 339217
## 12 member qui 323053
## 13 member sex 303302
## 14 member sáb 174712
Finding: As we can confirm, in terms of the number of rides made, members make significantly more rides everyday than casual riders.
DELIVERABLE: To sum up, there is a clear difference between casual users and members, whereas casual riders take fewer rides, but longer ones, which can suggest a more recreational use of the service; members have a much higher number of rides, but much shorter ones, which can imply a much more functional use of the service.
This is clearly shown and supported by the table below .
all_rides_v2 %>%
mutate(weekday = wday(started_at, label = TRUE)) %>%
group_by(member_casual, weekday) %>% #groups by usertype and weekday
summarise(number_of_rides = n(), average_duration = mean(ride_length)) %>%
arrange(member_casual, weekday)
## `summarise()` has grouped output by 'member_casual'. You can override using the
## `.groups` argument.
## # A tibble: 14 × 4
## # Groups: member_casual [2]
## member_casual weekday number_of_rides average_duration
## <chr> <ord> <int> <dbl>
## 1 casual dom 49154 36.5
## 2 casual seg 25719 23.0
## 3 casual ter 26391 25.6
## 4 casual qua 27666 24.7
## 5 casual qui 28303 30.8
## 6 casual sex 32142 28.2
## 7 casual sáb 44465 26.0
## 8 member dom 172335 10.0
## 9 member seg 323335 9.27
## 10 member ter 350426 9.03
## 11 member qua 339217 9.04
## 12 member qui 323053 9.00
## 13 member sex 303302 9.12
## 14 member sáb 174712 9.71
all_rides_v2 %>%
group_by(member_casual) %>%
summarise(average_duration = mean(ride_length)) %>%
arrange(member_casual) %>%
ggplot(aes( x = member_casual, y = average_duration, fill = member_casual)) +
geom_col(position = "dodge") +
labs(title = "Average ride time", subtitle = "per type of User", x = "Type of
User", y = "Duration of rides")+
scale_fill_discrete(name = "User")
2) Another very important finding, regarding the average ride time, is that this trend, of casual user riders being in average much longer than the member rides, holds across all days of the week.
As a reminder, here are the corresponding days in english:
Let’s go the visual
all_rides_v2 %>%
mutate(weekday = wday(started_at, label = TRUE)) %>% #creates weekday field wday()
group_by(member_casual,weekday) %>%
summarise(number_of_rides=n(), average_duration = mean(ride_length)) %>%
group_by(weekday) %>%
mutate(percent_average_duration = average_duration/sum(average_duration)) %>%
mutate(percent_average_duration = round(percent_average_duration*100)) %>%
ggplot(aes(x = weekday, y = average_duration, fill=member_casual)) +
geom_col(position="stack") +
geom_text(aes(label = paste0(percent_average_duration, "%")),
position = position_stack(vjust = 0.5)) +
labs(title = "Average ride times", subtitle = "per day of the week and per type of
User", x = "Days of the week", y = "Duration of rides")+
scale_fill_discrete(name = "User") +
theme(title = element_text(size = 13),
axis.title.x = element_text(size = 10),
axis.title.y = element_text(size = 12),
axis.text.x = element_text(size = 10),
axis.text.y = element_blank())
## `summarise()` has grouped output by 'member_casual'. You can override using the
## `.groups` argument.
3) In terms of the number of rides, on the other hand, members make significantly more rides in total than casual riders—precisely 89.47% more rides, like the data has shown.
Let’s get a visual of it.
all_rides_v2 %>%
group_by(member_casual) %>% # groups by type of user
summarise(number_of_rides = n()) %>% # calculates the number of rides
arrange(member_casual) %>%
mutate(percent_number_of_rides = number_of_rides/sum(number_of_rides)) %>% # calculates the percentages on the number of rides
mutate(percent_number_of_rides = round(percent_number_of_rides*100)) %>% # calculates the percentages on the number of rides
ggplot(aes(x = member_casual,y = number_of_rides, fill = member_casual))+
geom_col(position="dodge")+
geom_text(aes(label = paste0(percent_number_of_rides, "%")),
position = position_stack(vjust = 0.5))+ # Adds percentages to columns
labs(title = "Percentage of all rides", subtitle = "per type of User", x = "User", y = "Percentage of rides")+
scale_fill_discrete(name = "User") + # Changes the legend title
theme(title = element_text(size = 13),
axis.title.x = element_text(size = 12),
axis.title.y = element_text(size = 12),
axis.text.x = element_text(size = 10),
axis.text.y = element_blank())
4) Also, in terms of the number of rides made, members make significantly more rides, regardless the day of the week, than casual riders.
Let’s have a look.
all_rides_v2 %>%
mutate(weekday = wday(started_at, label = TRUE)) %>%
group_by(member_casual, weekday) %>% #groups by usertype and weekday
summarise(number_of_rides = n()) %>% # calculates the number of rides
group_by(weekday) %>%
mutate(percent_number_of_rides = number_of_rides/sum(number_of_rides)) %>%
mutate(percent_number_of_rides = round(percent_number_of_rides*100)) %>%
ggplot(aes(x = weekday, y = number_of_rides, fill= member_casual)) +
geom_col(position="stack") + # transforms our standard column graph into a stacked column graph
geom_text(aes(label = paste0(percent_number_of_rides, "%")),
position = position_stack(vjust = 0.5))+
labs(title = "Percentage of all rides", subtitle = "per day of the week and type of User", x = "User", y = "Percentage of rides")+
xlab("Days of the week") +
ylab("Percentage of rides") +
scale_fill_discrete(name = "User") + # Changes the legend title
theme(title = element_text(size = 13),
axis.title.x = element_text(size = 12),
axis.title.y = element_text(size = 12),
axis.text.x = element_text(size = 10),
axis.text.y = element_blank())
## `summarise()` has grouped output by 'member_casual'. You can override using the
## `.groups` argument.
5) There is a noticeable dip in the number of rides that members take on weekends, although they still constitute a much larger percentage when compared with casual users.This dip could offer an interesting clue revealed by the data, helping us trace the profile of this user type. Combining this information** with what the data has already shown us — that members, on average, take much shorter rides — could strongly indicate that this user group tends to use the service more on workdays, likely for their daily commute**.
So, the data has shown there is a clear difference between casual users and members:
Casual riders take fewer rides, but longer ones, which strongly suggests a more recreational use of the service;
Members have a much higher number of rides, but much shorter ones, which can imply a much more functional use of the service, such as a daily commute to work.
Now that the analysis is finished and we have obtained our results, it’s time to present our conclusions and recommendations based on your analysis, to the Cyclistic Executive Team and the Director of Marketing.
So based on this analysis, these are my recommendations:
1) Since the casual riders have fewer number of rides, but a much higher average of ride time, of 28min:56sec vs the 9min:22sec of the members average ride time, I would recommend to change the casual ride plan to include a “long ride fee”, for rides over 19 minutes, above which the ride fee would increase in 15%, at the 20min mark; in 20% at the 25min mark, and of 35% at the 30min mark.
2) This condition, naturally, would not be applicable in the member plan subscription, which would have no limits on the number of rides, nor on ride time;
3) To make the member subscription plan even more attractive to the casual riders, I would recommend to launch a campaign of 35% off the member subscription plan to casual riders that used the service more than 20 times in the last 2 months. This would be attractive to riders who already use the service frequently and are loyal customers who just need a little “nudge” to go for the member plan, at that same time that prevents that someone, who just used the service once, could get access to this special price deal;
4) Lastly, I would recommend a member-get-member type of deal, where for each friend a member gets to subscribe to member subscription plan, he would get a 15% cash back in a voucher for goods and services, and the friend a 15% off his or hers own member subscription plan.