Abstract:
CitiBike is a popular transportation tool in New York City and is widely used by people across all ages. This project is designed to find out among those CitiBike riders,  who have more usage of CitiBike on weekends over weekdays.  The main idea is to divide the riders into two age groups:above 30 years old and under(includes equal) 30 years old. By utlizing one month of NYC CitiBike data and  Null Hypothesis Significance Test, we can see younger generations who are under 30 years old are more prone to CitiBike on weekends. 
Introduction:
Citi Bike is the nation's largest bike share program, with 10,000 bikes and 600 stations across Manhattan, Brooklyn, Queens and Jersey City. It is a quick and affordable way to get around town and very popular in NYC area. Analyzing  users' activities is one of the most important ways to understand the business and social behaviors. And in general, people in the city below 30 years old are children, teenagers or singles , many of them are students or new starters in their careers. Meanwhile people over 30 years old might have families and stable jobs. My project is trying to find out which group uses CitiBike more on weekend for transportation.
Data:
The data used for this project is CitiBike monthly ridership dataset . And specifically,  the month of June 2016 dataset is used for analysis. It is provided by CitiBike Program, which can be accessed at their official website: https://www.citibikenyc.com/system-data,  and  https://s3.amazonaws.com/tripdata/index.html. The dataset contains columns of trip duration, location information and riders information. To focus on our question mentioned above in introduction part. Only the column of the riders' birth year is kept, all the rest of the columns are removed. Then riders by people who born over 30 years ago are group together and sum up, same with riders by people who born less than or equal to 30  years ago. Finally we plot these two groups' data into two figures, one is total quantity of each group's weekly ridership, the other is the each weekday's ridership fraction within their own groups. Note that for two groups data are plot into same figure for comparison.