DATA PROJECT 2: NYC NIGHTLIFE
RESULTS OF STUDY

NYU HOTSPOTS
The most popular destinations for NYU students are East Houston Street, Union Square, 6th Ave (between 10th Street and 22nd Street), and Chelsea. These regions are a product of the top three clusters calculated using k-means. Specific locations with a high number of markers include Katz's Delicatessen, Rockwood Music Hall, Hudson Bar and Books, etc. In terms of trends over time, it's pretty apparent playing with the layers that NYU students have been travelling farther from campus, both uptown and downtown. The mean trip distance for 2009, 2012, 2015, and 2018 has not changed much however.

COLUMBIA HOTSPOTS
The most popular spots for Columbia students are West 96th and Broadway, West 65th and Columbus, Columbus Circle, lower Chelsea, and West 79th and Broadway. 96th and Broadway was a particularly popular destination, likely due to the express metro lines at that station. Other key destinations include the Lincoln Centre for Performing Arts, AMC Lincoln Square, and Chelsea Market. It appears as though the destinations are more activity-based for Columbia students. It also appears as though the destinations were concentrated on the West side, but over time, the East side became more popular.

NYU VS. COLUMBIA: OTHER KEY METRICS
To extend this analysis, I also compared key metrics to uncover other differences between Columbia and NYU. The data shows that Columbia students, on average, travel further, have higher trip costs, and give higher tips. The difference was statistically significant (p < 0.005) for all three of those metrics. Interestingly, the number of passengers in each trip was not statistically significant. I presumed that Columbia students would have statistically higher car pooling rates since they were distanced from the most happening parts of NYC and would've liked to split the cost. Moreover, the activities would probably involve a group. One explanation for this finding could be that the passenger counts are reported by the driver, and the values are therefore not accurate.

OVERALL ANALYSIS
According to the cluster map of all entries, the most popular areas in NYC for students are: East Village, West Village, Greenwich Village, and West 96th and Broadway.

CONCLUSION
Limitations:
Only data for Halloween
Only Yellow Cab data (increasing use of ride-sharing)
Captures information for older people
​
Next Steps:
Use Azure/AWS for more extensive data analysis
Incorporate Uber, Lyft, and Via data
Analyse traffic and congestion trends
Look into daytime/weekend activity data