Universities that Never Sleep

Fall 2019 
GSAPP Planning Elective, Urban Informatics I, Columbia University
Instructor: Anthony Vanky
Group Work



Nightlife is an essential identity of New York City. In 2019, the city has its first report on nightlife economy released, showing a two percent annual growth rate in nightlife establishments, which outpaces the city’s overall economy between 2011 and 2016. College students, meanwhile, are identified as an important group of population who have consistently contributed to the demand for and prosperity of nightlife economies, and nightlife services and activities have further become a head stream of their interaction and creativity. In order to further explore nightlife services provided for students in the city, we’ve selected the two bestknown universities – Columbia University and New York University – in New York to evaluate and compare the number and diversity of nightlife choices enjoyed by their students.

Our Methods

Our data is acquired from Yelp through API calls on October 13. We searched for “nightlife” businesses within0.75 mile (1,200 meters) that is a 15-minute walking distance from centers of the two campuses (116th Street for Columbia, West 4th Street and Washington Square E for NYU), which delineates the schools’ neighborhoods that may serve the students more often. Data acquired from Yelp include business name, title, rating, price, and review languages, based on which we divide the businesses into four categories: 1) restaurant, 2) bar, 3) arts, venues, and activities, and 4) others, and then evaluated the multiplicity and diversity of
businesses with a rating of 3 or higher (to be qualified as a good choice) from three dimensions:

    1) number of choices under each category,
    2) price distribution, and
    3) linguistical diversity.

Finally, we end with an attempt to calculate a diversity index for each type of nightlife services.


Analysis

Due to its prominent location, NYU has far more nightlife services available for their students than Columbia does (842 vs. 76). The two pie charts below show composition of nightlife businesses around each university, which reveal similar patterns with restaurants and bars taking the majority.

We further looked into all the restaurants based on the style of cuisine and made the tree maps to represent the ranked quantity. It is shown that American food enjoy great popularity near both schools. Nonetheless, Columbia has greater percentage of Mexican and Latin American-style food, whereas NYU seems to be a great place for exotic dinner, such as Lebanese and Austrian food. Asian food, with the emerging fusion style, also accounts for a significant part to NYU’s cuisine diversity. Generally speaking, NYU are served with more and more diverse nightlife services both in terms of the number and the type.

Whether the services are provided at different prices and thus qualities, however, constitutes another dimension of diversity, which makes special sense for students. Through looking at price distribution for each category of services, we find that while the patterns are similar again, Columbia University tend to have
more cheap choices, especially for restaurants, indicated by a higher percentage of businesses with two ‘$’. Meanwhile, there is no business makred as “$$$$” around Columbia, suggesting again the overall affordability of the services. More expensive businesses and services can actually be found around NYU, with a higher
percentage of businesses with 3 “$” or more, yet there are also a number of 1’$’ bars and art venues which indicate places for a quick drink or entertainment in late night.

As two prominent universities in Manhattan, both universities enjoy reputation for its cultural diversity. In this case, we attempt to interpret it through the linguistic diversity in business reviews (i.e., the number of languages used). In order to make the comparison, we first filter out those businesses with reviews in over one language, and calculate its percentage among total businesses under each category. The result shows us that NYU has more linguistically diverse nightlife services in its surroundings with 46% of the restaurants, 38% of bars, and 27% of art venues having comments in multiple languages, and the figures for Columbia are respectively 35%, 27%, and 17%. Moreover, there are 3 bars around NYU which have 10 languages used. NYU, in this sense, has its superiority revealed in terms of the cultural and language diversity the students enjoy.

Furthermore, we made an attempt to look into a diversity index to give a quantitative measurement of diversity in nightlife services provided. We adopted the Shannon-Wiener Index, which is broadly used in ecology for biological diversity, and takes both evenness and abundance of certain species into consideration. The formula is presented as below:

The S is the abundance (i.e., the total number of business in this case) ; Pi is the relative evenness (i.e.,the percentage of certain type of business). The results are summarized in the table below, and it is found that NYU, again, has shown greater diversity in types of restaurants and art venues. The use of the index is rather experimental, but we believe the idea of using an indicator for nightlife services deserves further investigation.

Conclusion

Without much debate, New York University wins the game with significant advantage than Columbia University in all terms of the types of nightlife services, the price distribution, and the languages used. However, there are a number of constraints and limitations with our analysis. First, the delineation of the universities’neighborhood areas can be defined in other ways, especially for NYU which has multiple campus sites. Second, while investigating the pattern under each category, we exclude the “others“, because of the difficulty in defining and interpreting them. Some of the businesses under this category may include tobacco shops or deli shops which are not defined as a quality place for nightlife, and this refers the third point: the definition of “nightlife” businesses in Yelp API calls are actually unclear, which makes the results rather fuzzy. A closer examination on the data may be required.




Work Cited

    1. The Mayor’s Office of Media and Entertainment(2019). NYC’s Nightlife Economy Impact, Assets, and Opportunities.
Retrived from: https://www1.nyc.gov/assets/mome/pdf/NYC_Nightlife_Economic_Impact_Report_2019_digital.pdf

    2. DIVERSITY INDICES: SHANNON’S H AND E. (n.d.). Retrieved October 13, 2019, from http://www.tiem.utk.edu/~gross/bioed/bealsmodules/shannonDI.html.

Data retrived from:
https://api.yelp.com/v3/businesses/search target="_blank">https://api.yelp.com/v3/businesses/search
https://api.yelp.com/v3//businesses/{id}
https://api.yelp.com/v3/businesses/{id}/reviews