Where are the Data on NY's Homelessness?
Sousveillance Tool
Summary
For this sousveillance project, I chose to investigate data about homelessness in New York state. Living in New York, I often see many homeless people on the streets in the winter, and I always wondered if the government is doing enough to keep them safe. Upon researching for data, I realized that there is barely any homeless data available online. The only two data that I could find were from the Department of Homeless (DHS) that stores information about homeless individuals in their system and from the Continuum of Care (COC). Upon analyzing the data further, I noticed that the amount of homeless people sheltered recorded in both databases are not equal. Since COC collects data from regional and local planning bodies that coordinate housing for homeless individuals, it seems that there are other organizations involved in sheltering these individuals. However, there is no open data source available about these individual organizations. Which begs the question, “how do we know that these data are correct?” and “even with other organization’s help, why are there still so many unsheltered homeless individuals on the street?”.
Design Process
Finding The Right Data:
The two data sources that I used for my project is listed below:
- Department of Homelessness: https://data.cityofnewyork.us/Social-Services/DHS-Daily-Report/k46n-sa2m
- Continuum of Care: https://endhomelessness.org/homelessness-in-america/homelessness-statistics/state-of-homelessness-report/new-york/
Investigating the Data:
DHC’s API contains a large amount of information collected from 2013 to 2021. The database collects the amount of individuals in homeless shelters every week. However, the data collected by COC is an average number per year. So, the first thing I had to do was convert DHC’s data by averaging out the numbers of homeless individuals in shelters per year. My process is shown below:
nyData[year].dataPointCount += 1;
nyData[year].dataPointSum += parseInt(dataPoint.total_individuals_in_shelter);
nyData[year].average = nyData[year].dataPointSum / nyData[year].dataPointCount;
The next data piece I had to tackle was finding the difference between the average number of sheltered homeless individuals in shelters between the two databases. The formula I used is shown below:
COCData[yearIndex].Sheltered-nyData[years[yearIndex]].average
Designing the Data Visualization:
I thought of three designs initially, but decided to go with a bar chart as it will be easier for the viewer to understand the data presented. It is much harder to see the difference between the two sheltered homeless data with multiple pie charts. On the other hand, the line chart would not be as impactful to show the missing data points.
Debugging The Tooltip:
As I finalized my design, the biggest issue I had was getting the tooltip to work. My initial design, shown below, the if statements for mouseX and mouseY is nested in a for loop. Each for loop I had determines how and where each bar is drawn. So, if the if statement is nested in a for loop, once the other for loop is executed, it will draw on top of the initial for loop containing the if statement. This is a problem since everything is under the draw function. Hence, I had to execute the tooltip function only at the end of the draw function, to ensure my tooltip is shown on top of all the other elements being drawn.
To fix this, I stored the values of the rectangles in an array which I named rectangleData[] and wrote an if statement that will detect the mouse’s X and Y location. The method allowed me to draw my tooltip on top of the chart elements. The final result is shown above.
Reflection & Conclusion
What does my Sousveillance tool do?
I created this data visualization in hopes of raising awareness about how there is no transparency about the data collected on homeless individuals. Upon researching, I realized that only COC and DHS have open source data about homeless individuals in New York. Non-profit organizations do not seem to be giving out this data freely. This raises many concerns.
1. How do we know that this data collected, and shown, is reliable?
Without open data available, there’s no way for citizens to check whether the data is correct. There’s no transparency on which organizations are sheltering the homeless, or whether the data collected by COC is reliable. Perhaps, this is a bigger problem in our society. The issue of homelessness is often overlooked by those who are privileged. It is unfortunately normal for government bodies to ignore the importance of homelessness data as these citizens are not contributing to the economy. Or perhaps, the government is trying to hide something. Is it possible that they do not provide transparency because they do not want the country to know how bad the homeless crisis is?
2. What other organizations are sheltering the homeless?
COC’s website only stated that it collects data from regional and local planning bodies that coordinates sheltering efforts for the homeless. Without knowing the names of these other planning bodies, there’s no way for normal citizens to know whether the data collected represents the situation as a whole. The difference between the number of DHC sheltered homeless and COC sheltered homeless (in purple) is alarming. Is it possible that they neglect certain organizations in their data collection efforts? Are there actually more homeless people in shelters or on the streets?
3.Even with other organizations helping, why is government-run DHS still struggling to shelter all homeless individuals in NYC?
From 2013-2018, there’s an average of about 4000 homeless individuals who are unsheltered. That means, these individuals are left out in the cold during New York’s gruesome winter. Does this mean that the government has not allocated enough fundings to help them? Or does this pose a bigger question about the price of rent in New York?
4. Where is the most recent data? Why are there no more collected recently?
Only DHS contains the most updated data on its system. COC’s report only stops at the year 2018. I cannot find any other data about homelessness in New York at all, and that is very alarming. Is my initial hypothesis that the government body puts little effort into helping these individuals because they don’t contribute to the economy?
The lack of transparency on homelessness data concerns me. Although these individuals do not contribute to the economy due to their hardships, it does not mean that the government should completely overlook them. Afterall, they are still citizens of the United States. I hope my sousveillance tool can raise awareness on the lack of open data (as a whole), the lack of transparency on homeless data, and the alarming growth of homelessness in New York.