The Ford Go-Bike dataset contains 183412 records of bike rentals in the bay-area in February 2019 and divided between 16 columns. The dataset gives informations about the attributes of each trips such as the duration, the start and end station and about users such as the date of birth or gender and user types.
In the exploration, I found that there was a strong relationship between the duration of trips and weekdays, with modifying effects from start hour and user type. The duration of the trips is longer during the weekends the majority of Go users are subscribers but the most trips are made by customers who apparently often make trips at night the trips are mostly made by people aged between 20 and 40
Outside of the main variables of interest, I verified the relationship between user and type and bike share for all trip option. Few subscribers are taking this option.
For the presentation, I focus on just the influence of the user type and weekdays and leave out most of the intermediate derivations. I start by introducing the duration of trips.
Afterwards, I introduce each of the categorical variables one by one. To start, I use the bar plots of duration and weekdays, duration and start hour and then box plots of duration and user types.