Exploring the Value of Predictive Analytics for Strengthening Home Visiting

Evidence from Child First

By Samantha Xia, Zarni Htet, Kristin Porter, Meghan McCormick

Predictive analytics, or the use of historical data to forecast future outcomes, has long been a feature of business and marketing research. But increasingly, predictive analyt­ics is being applied in the social service and education domains. Program administrators can use relatively simple approaches, such as models with a limited number of measures, to predict outcomes of interest. Or they may consider more complex machine-learning models, taking advantage of the large amounts of data that organizations often collect, to improve their ability to make predictions. Either way, the promise of predictive analytics is to help programs identify those clients who could most benefit from targeted interven­tions—facilitating effective service delivery at an efficient cost.

An ongoing research partnership between MDRC and Child First, a home visiting pro­gram that provides therapeutic support and services to families with young children, offered the opportunity to examine the potential benefits, if any, of using predictive analytics to improve service delivery. To date, these methods have had limited applications in the home visiting domain. This brief offers results from that proof-of-concept exer­cise. Additionally, the brief provides much-needed information on the value of predictive analytics for similar organizations and may be a helpful guide for future researchers and practitioners, as more programs seek to implement these cutting-edge analytics tools.

Xia, Samantha, Zarni Htet, Kristin Porter, and Meghan McCormick. 2022. “Exploring the Value of Predictive Analytics for Strengthening Home Visiting.” New York: MDRC.