Very often, instead of studying at home, I go to one of the cafés in my neighbourhood. These are the places where I work on my assignments, respond to messages or simply enjoy coffee and life. What I have noticed is that usually, I’m not the only one! Therefore, I thought that the topic of sidewalk cafés can be interesting for everyone. In my research, I decided to map cafés to the train lines and train stations. I was wondering whether there is a bigger number of cafés next to train station entrances, which would be very useful in a rainy day in NYC (which happens rather often!). As an end product of my design, I wanted to create a map, which could be used by anybody who would like to have a coffee at a sidewalk café and would like to know which train they should take to get there.

My inspiration

In my visualization, I was inspired by the work of my colleague – WenLin Pan. I really liked the topic chosen by her and thought it was a great idea. WenLin at her work used Tableau to analyse and visualize the data. Interestingly, she also mapped her data, however she chose a choropleth and Zip code areas. This approach presents the areas with the biggest density of sidewalk cafés, however users won’t be able to find one they like and take the train to go there, which was something I wanted to achieve.

Software and used databases

To create my map, I used Carto as a software and 3 different datasets: 

• Sidewalk Café Licenses and Applications – available at NYC Open Data portal, last updated on November, 15th 2019
• 2016 (May) New York City Subway Routes – available at NYU Spatial Data Repository, last updated in 2016
• City Subway Stations – available at NYC Open Data portal, last updated on August, 22nd 2019

Methods

To achieve my aim and create my visualization, I have created 4 layers, where a map of NYC provided by Carto was the most basic one. All other data was just mapped on it. All databases, which I have used were tidy and I did not need to adjust them. In the Sidewalk cafes in NYC layer, I have performed an analysis and created Travel or Distance Buffer of 100 metres as I believed, that this is the distance which seems realistic for users to potentially run in the rain to the nearest café from the train station entrance. Additionally, I decided to color cafés in different shades of burgundy according to the number of chairs they offer. I thought that this distinction might help users to assess the size of a café. 

While deciding on color schemes used for this visualization, I wanted to focus on the clarity of my design and on esthetics. I rejected the idea of coloring different train lines, as the focus of this map is on the entrances to the train stations not necessary the lines themselves. Therefore, I have created labels for both train stations and cafés. User can click on the dot representing the train station and find out what is the name of the train station and which trains stop there. When users click on the café names, they receive the name of a café, address (street name, house number) and number of chairs. I have also chosen a map which has all green areas and streets marked. I assumed that this will be more natural for users. Additionally, I enabled the widget narrowing list of cafés by street names. It might help user to find the places they want to go to.

Results

While the final design decisions were guided by the purpose of the product – use by everyone who would like to come to NYC and have a coffee in a sidewalk café, I also wanted to investigate a few of my prior assumptions. 

I assumed that the entrances to the train stations are natural places for a sidewalk café – there is usually a bigger congestion of people, which everyone who lives in NYC could experience. However, this assumption appeared not to be correct. While looking at the map, it is visible, that there is no strong correlation between train entrances and cafés. 

The biggest number of sidewalk cafes is in Manhattan in the areas of Greenwich Village and East Village. Possibly, it is because of the way these areas are planned (narrow streets, lack of high car congestion). Other areas, which have a high number of streetwalk cafes is Hell’s Kitchen, Upper West Side and Upper East Side. Hell’s Kitchen is well known for its dining opportunities, and both Upper West Side and Upper East Side are mostly residential, rather wealthy areas.

Cafés in Brooklyn and Bronx are more spread out. Their biggest congestion is in Williamsburg. Surprisingly, this dataset does not contain any data about cafés in Staten Island. More investigation (preferably personal!) is required.

Future research directions

I believe that this project has a great potential and my work is just a beginning. Based on my data, it is possible to use R language to check whether my visual assumption that there is no big correlation between sidewalk cafes and entrances to the train station is correct. This could potentially lead to more detailed project. 

It would be interesting to investigate all cafés – not only with possibility to sit outside. Additionally, it would be interesting to check where popular chain cafes are situated and check if there are any interesting patterns and differences between sidewalk cafes, chain cafes among them, cafes without outside sitting and chain cafes among them. This data could be compared with available policies of big companies and interviews with owners of local establishments.

The printouts of more detailed version of this map could have been distributed among local community – just as it happens with bookstore maps during Indie Bookstore Day.