NCCD Now: Predictive Analytics in Human Services

Our Staff


NCCD Now: Predictive Analytics in Human Services

National Council on Crime and Delinquency

For organizations looking to use the best data-driven practices to aid in the delivery of human services, "predictive analytics" and its sister phrase, "big data," have lately achieved buzzword status. Many people may not know exactly how predictive analytics works, but they have seen its results on sites that tailor products and advertising to the user, including Netflix, Google Search, and Amazon to name a few.

Broadly stated, predictive analytics is a way of looking for patterns in past data to estimate future outcomes. Netflix knows what you watched and what other people with similar histories watched, so it uses that information to predict what you might like to watch next. As you've probably experienced, there are times that recommendation leads you to your new favorite show as well as times when it leads you to scratch your head and think, why would they ever think I'd like this?

It's easy to understand why a methodology that seems like it could predict the future might be exciting in the context of child protection, juvenile justice, or adult criminal justice. If there was a way to protect children from abuse and neglect, especially serious injuries and fatalities, we'd all support using it. In juvenile justice, agencies are looking for tools to prevent outcomes (like future arrest) for young people.

So what does predictive analytics offer for our nation's children, young people, and families, as well as the human service agencies that must make tough decisions about how to spend resources and target interventions?

This month's web feature aims to answer that question. To inform the discussion, we'll be bringing you blog posts on all aspects of predictive analytics: what is it? How does it work? How does using it interact with social work practice? What are the ethical considerations? What are its promises and limitations?

We'll also hope you'll join us for a webinar on September 3 featuring chief program officer Dr. Jesse Russell. Dr. Russell has recently published a paper on predictive analytics in child protection. The webinar will discuss the potential and constraints of predictive analytics in child welfare, and how NCCD is working with jurisdictions to explore their local needs and strengths around using this methodology.

Using data to improve human services practice and policy is NCCD's main focus. We are always looking for better ways to improve our systems, support workers, and help kids and families be successful and safe. We look forward to a lively conversation this month with you. Blog posts will be listed here as they go live.

Are You an Ethics Champion? 5 Questions to Ask Before Employing Predictive Analytics in Practice by Jesse Russell, PhD, Chief Program Officer

Data Talk: 5 Concepts About Predictive Analytics by Jesse Russell, PhD

Is Actuarial Risk Assessment Predictive Analytics? by Jesse Russell, PhD

Predictive Analytics and Child Protection Practice by Phil Decter, Associate Director

Case Example: Using Predictive Analytics to Answer a Child Protection Question by Colleen Kerwin, Researcher

The Consequences of Mistakes in Human Services Decision Making by Kristen Johnson, PhD, Senior Researcher

Fatality-Driven Child Protection Systems: Helpful or Harmful? by Raelene Freitag, PhD

A Recipe for Found Data by Andrea Bogie, Researcher

The Three W’s of Predictive Analytics by Tim Connell, PhD

Garbage In, Garbage Out: The Role of Existing Data in Predictive Analytics by Tim Connell, PhD

Who Should We Worry About? By Erin Wicke Dankert, Researcher


Predictive Analytics in Child Protection--Constraints and Opportunities by Jesse Russell, PhD

Three Examples of Predictive Analytics in Child Protection by Jesse Russell, PhD

Being a Good Consumer of Predictive Analytics in Child Protection by Jesse Russell, PhD

Submitted by Visitor on September 3, 2015 - 12:30pm.

Thank you

Add new comment

Filtered HTML

  • Allowed HTML tags: <a href hreflang> <img src alt height width> <em> <strong> <span> <cite> <code> <ul type> <ol start type> <li> <dl> <dt> <dd>
  • Missing filter. All text is removed
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Enter the characters shown in the image.