Introduction

Safe and sustainable public housing is a serious issue in New York City that was even more prominent in the COVID 19 pandemic. I hope to learn more about how public housing and public transportation work together, or do not work together through a series of visualizations supported by the datasets from NYC Open data like this one. Some questions that I will consider:

• How close are the housing developments to subway entrances?
• Which zip codes can to create more housing developments near subway entrances?
• How can this map be a resource for NYCHA?

Resources like the NYC Equity Indicators help provide context for this query through investigation of accessibility to public transit. Scores are given, by year, to assess factors that effect New Yorkers commute times. In their comparison of race and commuting times they report, “Long commutes have been linked to compromised physical and mental health and lower life satisfaction. In the US, racial and ethnic minorities, lower-wage and lower-skill workers, and people who live in high-poverty communities typically have longer commutes.”

In order for the New York City Housing Authority (NYCHA) to provide housing that supports better commute times and overall life satisfaction, consideration must be made of where housing developments are built relative to their access to public transportation. And as the need for more affordable housing in New York City continues to rise, more developments will need to be constructed.

My goal with this data visualization is to create a resource map for NYCHA users showing clear relationships between public transit access, specifically subways, and public housing developments by zip code acting as a tool to determine where developments could be placed to improve the quality of New Yorker’s commutes.

Process

I chose to review the current resource maps NYCHA has available to the public. The NYCHA development introduction says that anyone can, “use our interactive mapping tool and online directories to get key details about any development, such as its address, resident association info, on-site resources and facilities, photos, maps, demographics, and more. You can even find the development’s closest subway, school, or library.” I found this an exciting place to start to take note of my thoughts about each resource, and below is a description of that process.

This map contains lists of all the developments throughout the five boroughs. Each point is color coded by the housing program type. The legend on the right of the map has the name of every development and its index number. The boroughs are clearly labeled and the color scheme is extremely clear. However, I find the text very inaccessible in its size and the amount of it can be overwhelming as demonstrated in the direct labels seen here.

After assuming that this map would include more of the resources listed in the introduction, I visited another resource shared on the developments page. Below is an interactive map that allows the user to search developments by boundary. The landing page of the map loaded quickly, and on the right as seen here there is a large “layer list” legend corresponding to boundaries on the map. The attribute layers can be turned on, collapsed or closed completely. In this list users can also open an attribute table with more data. All of the filters show government run initiatives like Neighborhood Rat Reduction or government run facilities like police precincts and senior community centers. Boundaries of as assembly, senate, congress and council representation are present as well.

I have yet to find a map that clearly represents the relationship between public housing developments and public transit. Yes, in the example above the base maps are adjustable and if you know the city well it can be inferred where they relatively are, but there is not clear relationship. There is an interactive art map and a development map per borough by development. I try this last option. This is the map I found for a development in Williamsburg, Brooklyn.

This map gives a detailed view of the building and the streets and parks surrounding the development. It does not however show proximity to the public transit. So after reviewing all of these resources, I asked two participants to also navigate the NYCHA website to see if they would be able to find out how close a development is to a subway entrance or where more developments could be built to improve New Yorker’s commutes.

The first participant is a 36 year old male working as a union electrician, very familiar with subway construction and routes. He also identifies as color-blind. This is an important perspective for my research because he has expert knowledge of transportation and he will share an opinion about color that is not always considered while making data visualizations. He used my laptop to navigate the same NYCHA development page that I tried to navigate. He clicked through three maps and said he could not answer my questions without “cheating” by just referring to his phone to see how close the development addresses were to the subways. He also indicated that many of the colors were difficult for him to decipher especially in the smaller icons. The second participant is a 35 year old male living in a public housing development in Harlem. He was interested in my research and expressed because he wanted to “put his attitude to the test” in regards to his overall frustration with NYCHA. He went through and clicked on only one map, read for the word “subway” and asked me where he was supposed to click. Neither of the participants were able to answer the two questions, and this result left me with the same concerns.

My experience with the NYCHA website and the feedback from two users informed my own design of a specific resource to measure the amount of public housing developments near subway entrances.

Rationale

I chose to focus on these specific factors for three reasons. 1) When looking for a place to live, many New Yorkers are thinking about “what stop” they will be near referring to the subway, not necessarily which bus route. 2) City bike is a great initiative to make it easier to get to subway stops, but not everyone is capable or physically able to ride a bike. 3) Both Public housing developments and subway systems are large infrastructure and cannot easily be moved or rebuilt, so their presence, once created, it relatively fixed. With those in factors in mind, I chose to create boundaries by zip code as a comparison tool for count.

I chose to use Carto Builder for this visualization. Because I am layering multiple shapefiles this platform seemed to best fit my data needs. This is what the resource map looked like after I uploaded the sources and removed the fill color on the zip code boundaries to better see datapoints.

I was noticing a lot of overlap and I couldn’t make out what the correlation was between the subway entrances and the public housing developments. I initially also thought that the GoogleMaps basemap was going to be a good idea since I was visualizing transit; however, it was distracting and not helpful for the main research questions.

I removed the GoogleMaps base map and went with a dark background with a color scheme visible for color blind users. I wanted a better way to visualize the distribution of housing developments so I tried a heat map. This did not work. A heat map can have negative connotation, and I do not want to conflate that with a map of public housing developments. So I chose to use a light basemap and a cluster instead.

Along with a title, this version represents the aesthetic solutions to the previously listed problems. The Clustering is represented by three sizes, chosen through proportion during zooming in and out of the visualization to maintain the least amount of visual clutter and with a color visible for color blind users. The emphasis is given to the black dots representing subway entrances. This is a detail of the resource map.

Findings

As I noticed before in all the overlap, there is a strong correlation between subway entrance and public housing development. This is true across borough. There are places that have more density than others and this can be compared by highlighting a zip code and comparing its data to another, here is one comparison.

• 

• 

The image on the left represents a place that could potentially have more public housing developments near subway entrances, and the image on the right represents a place with a higher density of public housing developments near subway entrances.

This could be a resource for NYCHA because it makes clear relationships between these three factors. Instead of filtering through many other components, this resource map has a more focused intention and minimal design.

The Subway station points can be interacted with as well. After clicking on them, a link to the MTA website pops up as well as the lines the station services and the name of the station. This can be helpful when thinking about surrounding architecture and commute.

In response to user feedback, I made sure to make minimal, clear aesthetic choices. I understand that the user feedback from the two participants was most likely influenced by their experience using the NYCHA website. However, this was my experience as well. And I chose to have two rounds of user feedback, one before creating my own visualization and one afterwards, to recreate the order of my own experience.

NYCHA has committed to a Blueprint for Change NYCHA and they have also signed a joint letter urging Congress to fully fund NYCHA’s public housing infrastructure, modernization and capital projects for the first time in 40 years against the backdrop of a $40 billion deficit. Whether this leads to increased development is still uncertain. The Transparency and Reports page lists reports from each year detailing the ways in which developments are growing and changing.

Recommendations

It was not until I presented my visualization to the class that I realized that one of my datasets, NYC subway entrances, does not include the Staten Island Railway. I was glad that it was pointed out to me, and I was eager to patch in that shapefile to create a more complete resource. However, in my research, I was not able to find a shapefile for the Staten Island Railway. I though that I could maybe approximate the entrances on the map by adding points, but that did not end up working out. Then I though about removing Staten Island from the visualization, but I decided that would be even less complete and leave out an entire borough of data. The solution is probably available to me somewhere, and if I find it, I will certainly add it.

Another recommendation I have for future iteration is to use feedback from a user who works with NYCHA. Since their website was such a large part of my research, I think that it would be great to have an inside opinion about what I have created. They also may be able to help me understand the page.

I also think that the interactive map on the NYCHA website powered by Esri is a very elegant platform for this kind of visualization. Had I taken the time to learn more about it, I may have decided to use it instead of the Carto Builder platform. I say this because the customizability of the layers for users along with the basemap choices makes interacting with the map a better experience. I was able to filter and move through very easily.