Predicting less, understanding more
Shifting the use of machine learning from individual prediction to structural insights in systems that affect children and families
Bibliographic Data
| ID | 22414027 |
|---|---|
| Authors | Amanda J Fairchild (0000-0001-8668-4658, University of South Carolina, corresponding author), Josh Gupta‐Kagan (0000-0003-0783-9469, Columbia University), A Barclay (0009-0008-9483-428X) |
| Year | 2026 |
| Volume | 188 |
| Pages | 109120 |
| Publication date | 2026-09-01 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Children and Youth Services Review (JOURNAL) |
| Journal identifiers | ISSN: 0190-7409 • E-ISSN: 1873-7765 |
| Publisher | Elsevier BV (PUBLISHER) |
| DOI | 10.1016/j.childyouth.2026.109120 |
| OpenAlex | W4410942640 |
| Language | EN |
| References cited | 49 |
Machine learning algorithms (MLAs) have been deeply embedded across domains, valued for their capacity to analyze large-scale data and their ability to support a range of descriptive and predictive tasks. Despite their versatility, most applications of MLAs to date have focused on individual-level prediction at the expense of broader structural insights. This paper shifts that focus by using MLAs to examine how juvenile and family court intake structures shape case outcomes in delinquency referrals. Specifically, we apply machine learning to evaluate how the structure of juvenile court intake decision-making affects juvenile referral outcomes. In doing so, we offer a novel demonstration of how these tools can support causal inquiry into the functioning of complex legal systems. Results indicate that juvenile court intake structure has a material effect on youth outcomes in the system. That is, when prosecutors act without input from the juvenile justice agency, youth referrals are significantly more likely to be prosecuted rather than diverted or dismissed. Differences in prosecution rates between prosecutor-controlled intake (PCI) and agency-involved intake (AII) structures are most pronounced among Black girls and least pronounced among White boys, highlighting how system design choices may contribute to racialized and gendered disparities in juvenile justice outcomes. Taken together, these findings underscore the potential of MLAs not just to predict individual outcomes, but to generate insight into the policies and practices that shape them—opening new directions for structural analysis and reform in systems that affect children and families.
Cognitive psychology · Machine learning · Cognitive Abilities and Testing · Communication · Computer Science · Early Childhood Education and Development · Psychology · Socioeconomic and Demographic Analysis · Artificial Intelligence
Regression Modeling Strategies
On the practice of dichotomization of quantitative variables.
The Cost of Dichotomization
Dichotomizing continuous predictors in multiple regression
Contexts of Decision Making in the Juvenile Justice System
Justice by Geography
The Effect of Youth Diversion Programs on Recidivism
The cost of dichotomising continuous variables
Causal Inference Using Potential Outcomes
Who Will Graduate? Disruption of High School Education by Arrest and Court Involvement
Iatrogenic effect of juvenile justice
The meaning and use of the area under a receiver operating characteristic (ROC) curve.
Identification of Causal Effects Using Instrumental Variables
Estimating causal effects of treatments in randomized and nonrandomized studies.
Statistics and Causal Inference
Unfolding Algorithms
Why Children Follow Rules
Tool for Surveillance or Spotlight on Inequality? Big Data and the Law
Exploring Disproportionate Minority Contact in the Juvenile Justice System Over the Year Following First Arrest
Collateral Consequences of Imprisonment for Children, Communities, and Prisoners
Structured decision making in juvenile justice
Operationalizing intake
Race, Drugs, and Policing
Why Propensity Scores Should Not Be Used for Matching
Juvenile Arrest and Collateral Educational Damage in the Transition to Adulthood
| Highly cited | No |
|---|