The study — conducted by the data science firm Civis Analytics and the nonprofit food benefits enrollment advocacy group mRelief — was conducted in Los Angeles County from January to April of this year. It was designed to test a pair of potential improvements. The first was the ability to schedule a call directly with the CalFresh office, which handles food benefits enrollment in California. The second was the ability to schedule a call along with a text reminder to schedule a call. The study was conducted via a randomized control trial that ultimately included about 2,300 people.
What the research found was an 18 percent increase in enrollment within the group that was given the chance to schedule a call. Subsequently, text reminders showed no increase of any significance.
During a recent phone conversation with Government Technology, mRelief Executive Director and Co-founder Rose Afriyie attributed the increase to removing the burden of who has to initiate contact. Normally, eligible food benefit recipients might call for information and be told they have to call the CalFresh office for food info. What the study did was offer to essentially just put them through to that office.
Increasing the number of eligible people who sign up for food benefits has become a major focus of national civic tech efforts, especially in the nation’s most populous state. In fact, the national nonprofit and nonpartisan civic tech organization Code for America (CfA) recently announced it would take its own benefit enrollment increase effort, GetCalFresh, statewide.
Afriyie said this study is complimentary to what CfA is doing. CfA’s efforts are focused on online enrollment, while this study is concerned with how to get people who primarily interact with government via phone enrolled, especially those who lack necessary skills to enroll through other entry points.
“For us this is about a solution that empowers people who may not have the literacy to navigate an application,” Afriyie said.
This work won’t stop with California, either. Now that the study has helped validate a human-centered design tweak that gets results in that state, mRelief is piloting similar programs of a smaller scale in states like Arizona and Idaho. The organization’s mission is to work nationwide to ensure that an additional 1 million Americans who are eligible for food assistance actually receive it. Once it has done that, Afriyie said the group may look to take what it’s learned and apply it to other adjacent verticals like health care and housing.
Sherry Shenker, a data scientist with Civis who worked on the study, said being able to prove with data that an 18 percent rise is possible with such a small change is vital to fostering governmental buy-in. It’s one thing to discuss lofty ideas like human-centered design and another to have demonstrable numbers.
