AUG 04, 2022
5 lessons on data science, storytelling and place from the Delaware Data Innovation Lab
JAN 13, 2022
What’s the relationship between data and storytelling?
We discussed that and more in Technical.ly’s latest live Slack interview, featuring six members of the Delaware Data Innovation Lab (DDIL) team: Associate Directors Ryan Harrington and Héctor Maldonado-Reis, and fellows Jameelah Young, Nami Sunami, Mohammad Baksh and Mehak Gupta.
The team of data scientists, now working under Tech Impact, is on the heels a couple of major New Castle County CARES Act-funded project deadlines and are heading into 2022 with new and continuing community-focused work.
If you don’t know much about the lab, Harrington described it like this:
“DDIL is an organization focused on using data for social good. We were formed about one year ago during COVID. It came from observing how many insecurities our communities were facing — food insecurity, housing insecurity, economic instability. We believed that we could augment the work that professionals in those areas were doing by applying different data techniques to the problems. That’s still true today!”
With six AMAers, the hour sped by, with multiple threads on several topics going at once. You can read the whole convo on our public Slack in the #ama channel.
Check out some highlights below; some responses have been edited for clarity.
When it comes to that data and storytelling question, the DDIL crew noted that one helps us understand the other through a more human lens.
Sunami, DDIL’s most recently added fellow, said a narrative is important, if not imperative, to data science.
“Without a story, data/insight won’t be effective in many cases,” he said. “A story without data leads us to a very dangerous path. We do need both to achieve both knowledge and an effective change.”
One of the ongoing projects is with the Delaware Housing Authority, where data science is used to investigate the huge 75% discrepancy between Delawareans who need housing aid and those who actually receive it.
“DDIL is more so focused on surfacing data/insights that will help end-users make informed decisions as they address these sorts of gaps,” said Young, one of the fellows. “As we iterate on our projects, we learn more about exactly who the end-users are, and how we can best create data products that enable them to make informed decisions. End-users on this issue may be housing advocates and organizers, policymakers, researchers, etc.”
One of DDIL’s projects is the Energy Efficiency Index and map, which aims to geographically determine the need for more energy efficient practices using data from public and private sources, including the Department of Energy and electricity providers.
Fellow Baksh said factors such as age of the house and median income play an important role in estimating energy efficacy.
“For example, if a person lives in an older home and the home is located in an area where median income is 10 to 20% below the average median income, it would be expected to have older and less efficient appliances which will result in higher energy consumption,” he said. The size of a house or apartment and materials used to build the home may also be factors.
When asked what’s interesting about Delaware as a place to focus research, Harrington noted its small size and interconnectedness as benefits.
“Delaware is amazing because of how interconnected it is,” the associate director said. “It provides an opportunity to quickly and easily meet with all of the stakeholders that you need for success. The leaders in the state are easy to reach and easy to talk with. That makes a real difference for projects like this. Delaware’s size and interconnectedness gives it the chance to be a ‘lab’ state — place to work towards addressing some of the problems that we care about and then scale them to other places facing similar challenges down the line.”
“Also, Delaware is composed of less than 1 million folks at the state level — that is less persons than in Philly alone!” added DDIL’s other associate director, Maldonado-Reis. “The population and state size allow us to explore and deploy meaningful tools with a targeted approach.”
Especially during the pandemic, we see data about COVID-19 cases every day, but the impact of data science goes well beyond tracking large-scale health events. A project DDIL is doing with health system ChristianaCare, for example, targets other chronic conditions.
“The report is about distribution of chronic diseases and utilization of different care settings in state of Delaware,” said Gupta, a fellow. “This report will be helpful to determine the prevalence of chronic disease at census tract level and the need of different health care settings in Delaware.”
If you’re interested in seeing an actual DDIL partnership report, check out HCCD Innovation & Telehealth DHINsights exploring health claims data.
Natural Language Processing