This column is written for those interested in learning about the world of Data for Good, where statistical analysis is dedicated to good causes that benefit our lives, our communities, and our world. If you would like to know more or have ideas for articles, contact David Corliss.
With a PhD in statistical astrophysics, David Corliss leads a data science team at Fiat Chrysler. He serves on the steering committee for the Conference on Statistical Practice and is the founder of Peace-Work, a volunteer cooperative of statisticians and data scientists providing analytic support for charitable groups and applying statistical methods in issue-driven advocacy.
The holidays can bring warm memories of time with dear friends and family celebrating together. However, for those who struggle with housing insecurity and homelessness, this time of year can be fraught with even greater dislocation and difficulty.
Homelessness is an area in which analytics is much needed but often in short supply. Shelters and other organizations providing direct services to individuals many times have good data but no one to analyze it. Though they are making great strides to help, they could do even more with statistical support. The rapidly changing face of homelessness in America today is one of the main reasons analysis is so important.
Yesterday’s experiences, perspectives, and solutions may not work. As one example, Utah began a major push to reduce homelessness in 2005. The state’s efforts were effective at reducing chronic homelessness, which constituted the bulk of cases there in 2005. Today, however, the factors driving homelessness—such as affordable housing—have become much more prominent and homeless numbers are rising again. New statistical analysis at a local level is needed to keep up with these changes and to develop the best solution for each community.
Hackathons are one of the best ways to become involved in fighting homelessness with analytics. A quick web search will turn up many local events meant to develop homelessness solutions. In May, I had the opportunity to attend one sponsored by Data for Good / Toronto. D4G partnered with Youth Without Shelter, a local homeless shelter in Toronto with a variety of programs. The event was hosted by the Toronto office of Capco, a global management and technology consultancy. This particular hackathon focused on supporting Youth Without Shelter by helping them find more donors and volunteers. More than 20 teams spent the day working on different approaches, including marketing, geographic segmentation and analysis, and development of donors and volunteers over time. The entire program was a case study in how to run a good hackathon:
- Put together a small, core team of volunteers to develop and manage the event
- Identify an area of interest shared by the event team and a strong local organization working on the problem
- Arrange a corporate sponsor for the event that provides working space, restrooms, and coffee
- Arrange onsite participation during the event from the sponsor and organization being supported
- Register with an event management group (in this case, Meetup.com, which has good instructions for setting up an event)
- Promote the hackathon effectively via social media
Some academic statisticians, data scientists, and students may find today’s homelessness problem close at hand. A national survey from the Hope Center for College, Community, and Justice reports an increasing number of college students facing housing insecurity and as many as one in six homeless at some point in the past year. It’s a situation I can speak to personally from my student days, having gone from a dorm room to being homeless for seven months when just one semester short of finishing my bachelor’s degrees. It makes me wonder just how many people reading this article right now are homeless college students. Sadly, it is fairly easy to become involved with an organization where we live, work, or study
While many Data For Good advocates fight homelessness by partnering with an agency providing direct services to individuals, others study the underlying conditions driving homelessness to develop more effective policies and programs. Federal government resources are available through Data.gov. The National Center for Homeless Education also has data and other resources. And the New York City Housing Data Coalition is another great resource, compiling several publicly available data sources to support housing research. It is a particularly good place to look for data on the impact of housing prices and gentrification on new homelessness. The Data Challenge event judged at JSM this year used NYC housing data to produce a treasure trove of research, insights, and action plans for combating homelessness with data and analysis.
Whether through a hackathon supporting a local shelter, reaching out to a specific at-risk community, or doing basic research on the root causes of homelessness today and how to address it, Data for Good statisticians and data scientists are working together to find and develop new housing solutions. These efforts are making a difference in the lives of countless people to bring them home for the holidays—and every day.
As the year draws to a close, this is a good time to make plans for 2020. Think about your next D4G project, discuss it with friends and colleagues, and start putting it in motion. Working together, we can make 2020 another great year of making a difference in peoples’ lives through Data for Good.
“The Data Challenge event judged at JSM this year used NYC housing data to produce a treasure trove of research, insights, and action plans for combating homelessness with data and analysis.”
Were these research results and plans published somewhere? I would appreciate a reference or direction to a website.