The Number of Homeless Who Slept in the NYC Subway Increased Year Over Year.
There is no denying it, homelessness is a huge issue in New York City. Every traveler and resident that passes through the city will encounter many people living on the rat infested streets. It is so apparent that it is even talked about in songs about New York, such as Transit Ride by Guru. In which he says “Homeless people everywhere holding cups for changes, and if you have none then they look at you strange”. According to an organization called Coalition for the Homeless, in recent years the level of people who are homeless in New York City is the highest it’s been since the Great Depression. The problem has gotten so bad that in 2019 the New York City Mayor was caught secretly shipping homeless people to other places in the country, sent with them was a years worth of rent.
While the city has made great strides to help the less fortunate, it has not been enough. According to The Wall Street Journal, Mayor Bill de Blasio brags about providing beds in shelters for about 60,000 people. Within that article it is also revealed that over 3,700 people are still living on the streets and sleeping among the elements. For the purpose of this analysis I will be focusing on those who do not seek arrangements in the New York City shelter system. I want to analyze trends over the years for the unsheltered homeless people. In order to do this I found predictions from NYC Open Data Catalog of unsheltered homeless in each borough from 2009 to 2012. This data also included details about whether the people were living on the surface or in the subway.
In order to familiarize myself with the data I conducted some initial analysis. When looking at the total unsheltered individuals over the years it seems that 2009 is a low point, they spike in 2010, dip in 2011 and then spike again in 2012. This prompted me to look into if this is a theme present across all boroughs.
When looking at this visualization it’s interesting to note that this trend is the most apparent for Manhattan, slightly present in Brooklyn, and the least present in the Bronx. However, for the rest of the boroughs the number of homeless spiked in 2010 and then slightly decreased every year following. This also shows us that the city of Manhattan has the most unsheltered homeless people by a lot. When looking into this deeper I found that across all four years 56% of all unsheltered homeless reside on the streets of Manhattan. I am assuming this is because of the large amount of tourist and even resident foot traffic that passes through the center island. This gives them the best opportunity to beg for money.
When looking at the raw data there are surprisingly high numbers of people who seek refuge in the New York City Subway System. I wanted to examine this further so I pulled out the number of people who slept in the subway for each year.
Once I turned this into a line graph I found that every single year the number of people who sleep in the subway increases, on average about 222 per year. This is a scary statistic due to the fact the subways are already extremely crowded and are not meant to be lived in. This could explain why a lot of the subway cars and stations smell bad.
A few weeks ago following an interview in Long Island I met up with my brother in Manhattan. We went to a comedy club and I overheard the person behind me bragging about the amount of trees his neighborhood. Trees should not be something one should be able to brag about, but in the grey concrete jungle of New York City it is. This prompted me to look and see if the city has any tree data. Sure enough the city conducted a tree census and posted the data in 2015. I was intrigued by this and wanted to see if the amount of trees in a borough has any correlation to the number of unsheltered homeless.
There were over 680,000 trees captured in the dataset. To start I filtered out all the ones that were “dead” or “stump” under status. After doing this it reduced the list to around 650,000 trees. After that I broke them down into each borough and put those numbers in a new dataframe alongside the appropriate average unsheltered homeless.
I assumed that the more trees a borough had the more unsheltered homeless people it would have, given the homeless could seek shelter under the trees. This scatter plot tells a different story. While there is no definite correlation between the two it is interesting to note Manhattan with the highest number of homeless has the least amount of trees. While Queens with the lowest number of homeless has the highest count of trees. To see if this was a connection across all boroughs I conducted regression analysis. The correlation coefficient came out to being a .2, which shows a very weak positive connection. This is consistent with the scatter plot.
There were a few limitations to my analysis of the homeless in New York City. The first and most major one being the data was quite outdated. The city only released the raw data starting in 2009 and stopped in 2012. In recent years they only released graphs of the homeless that utilize the city shelters. The tree data and the homeless data is also from different years, which could skew the results. Another limitation of the data is that there is no reliable means of measurement to quantify the number of unsheltered homeless people. The number of people who sleep on the street changes everyday, yet they only supply one number per borough for the entire year. They also give a number of people who slept in the subway, yet how do they know these people haven’t been counted already at a previous subway station. This data is very important yet very hard to accurately acquire.
