Prediction as Prevention, with Emily Putnam-Hornstein
Can big data help us determine which children are most at risk of foster care placement? And how do we direct resources to those children to ensure they’re safe? We examine the way in which predictive modeling sheds light on the impact of implicit bias in our nation’s child welfare system. About 50% of African-American and Black families in this country will experience a child welfare investigation. That’s far, far more than the data indicates we should expect to see. That’s a problem. But can an algorithm be the answer? Emily Putnam-Hornstein, an associate professor at the University of Southern California School of Social Work and the director of Children’s Data Network, joins us to talk about what role big data should have in making potentially life-and-death decisions about children’s safety.
Topics in this episode:
- What is predictive analytics and how it is used in child welfare? (1:56)
- The big question to answered by big data. (3:52)
- The over-representation of black families in child welfare investigations. (5:31)
- Who gets reported? (6:58)
- Why haven’t we solved this problem yet? (10:01)
- Can individuals accurately assess risk? (12:24)
- How can predictive analytics address implicit bias? (15:24)
- How does it work in practice? (19:38)
- The impact of predictive analytics. (23:58)
- What’s next for the field? (28:48)
- Our next episode topic. (32:20)
“Can big data help prevent child abuse and neglect?” by Giles Bruce at the USC Annenberg Center for Health Journalism, talks about Emily Putnam-Hornstein’s work (June 24, 2019).
Transcript to come.