One of the best things about working at Civis Analytics is the number of opportunities we have to give back to the Chicago community. On September 23rd, Civis was proud to be a sponsor and participate in the Greater Chicago Food Depository’s Hackathon to End Hunger. The Hackathon is an annual event that brings the Chicago tech community together to address problems regarding food insecurity in Cook County, where 1 out of 6 people do not know where their next meal will come from. Last year, the Greater Chicago Food Depository distributed an average of 200,000 pounds of food every day to those in need, so the importance of leveraging great technology and tools cannot be understated.
This year, the hackathon had two focus areas, which stem from immediate issues the GCFD faces daily: (1) capturing food insecurity rates at the neighborhood level in real time; and (2) tracking how access to nutritious food impacts clients’ health.
Our group of Civis participants worked on the first focus area. Currently, the GCFD relies on a national model of food insecurity built on the American Community Survey to estimate food insecurity rates in Cook County. However, this model is based off of old data (circa 2015), and misrepresents food insecurity rates in specific populations. We broke the problem down into two questions: how can we survey the Cook County population to gather real-time food insecurity data, and how can we use publicly available data to better understand food insecurity in Cook County?
To answer the first question, some of us worked with other participants to design a short SMS-based survey to assess food insecurity rates by neighborhood. Although simple, SMS has fewer barriers than a traditional mobile app, so it can capture more accurate data among more vulnerable populations.
Other Civis participants worked on the second question, collecting a variety of public data sets from local and federal sources (the City of Chicago Data Portal, National School Lunch Program). We utilized MapBox to overlay data points — like the locations of all grocery stores, bodegas, and convenience stores, and how many students receive reduced-price lunches — on a map of Chicago. Using all these tools, we proved out the GCFD’s hypothesis that the ACS model is underestimating food insecurity in certain Chicagoland neighborhoods and showed how other publicly available data sources can be used to build a more localized and more accurate prediction of food insecurity in Cook County.
The map on the right is a heat map of food insecurity rates as determined by the American Community Survey (ACS). The map on the left is a heat map of food insecurity using the number of students, broken out by neighborhood, eligible for reduced or free meals through the National School Lunch Program. While the map on the left is not a complete picture of food insecurity, it helps prove the point that the ACS model does not fully capture food insecurity in certain neighborhoods and that there may be other factors, specific to Chicago, that can be captured through other publicly available data sets.
In the evening, all the teams gave five-minute presentations about what they built. Many teams thought of innovative ways to collect accurate information about food insecurity, like using kiosks or vending machines in public places. Other teams built dashboarding apps that allow doctors to visualize the health of patients along with the groceries they received at food pantries.
The judges, a group of Chicago tech leaders and a director at the GCFD, ultimately rewarded the top prize to the group working on the SMS survey. In their closing remarks, the judges appreciated the feasibility of the idea and novelty of using a chatbot to automate responses. In the end, it wasn’t about winners or losers but about bringing the Chicago tech community together to solve a pressing problem. The Hackathon to End Hunger reaffirmed our belief that technology can and should be used for social good.
Check out this video for a recap of the event, and please check out the Greater Food Depository of Chicago’s website for ways you can fight food hunger in our community. You may even see a Civis volunteer team at one of their food pantries!
To learn more about how data scientists are helping increase participation in the Supplemental Nutrition Assistance Program (SNAP), read our colleague Scarlett’s post about how the Robin Hood Foundation is using data to close program participation gaps in New York.
This post was co-authored by Ana C. and Bryan B.