Youth of color — Black youth most consistently — are overrepresented a

Loading the library…

For the Court

Criminologists agree on the pattern and disagree on the decomposition. The pattern: minority youth, Black youth most consistently, are overrepresented from arrest through confinement, and controlled analyses repeatedly find race effects surviving offense-and-record adjustment — largest and most consistent at detention and the punitive deep end. The decomposition: differential involvement explains part of raw disparity, differential selection is directly evidenced where offending is measured (Black first-arrestees offending less yet processed more), and the field's honest position, stated in its canonical review, is that fully apportioning the two may be impossible.

What This Claim Does NOT Say

That every disparity statistic proves discrimination — raw overrepresentation partly reflects differences in offending and prior record, which is why the controlled and matched estimates carry the weight.

That controlling for prior record fully cleanses the comparison — prior record itself accumulates through earlier system decisions, so controlled estimates can understate disparity (and unmeasured offending differences can inflate it in the other direction).

That any individual decision-maker's ruling was racially biased — these are aggregate patterns, not case-level findings.

That disparities are uniform — magnitude varies substantially across jurisdictions and decision points.

That reform mandates have resolved the problem — evaluations of the federal DMC mandate find disparities largely persistent.

Scope — Where It Holds

Applies to U.S. juvenile justice processing decisions (arrest referral, diversion/petition, detention, adjudication, disposition, waiver) measured in administrative and cohort data across four decades. Disparities net of legal factors are most consistent at earlier and deep-end decisions and vary by jurisdiction and decision point; Black-White comparisons are the most studied, with Latino and Indigenous disparities documented but thinner.

Sources & Verification

Seminal

Zane SN, Pupo JA (2021). Disproportionate Minority Contact in the Juvenile Justice System: A Systematic Review and Meta-Analysis. Justice Quarterly.
Meta-analysis · justice-processed juveniles (U.S.)
Systematic review and meta-analysis of the disproportionate-minority-contact literature: synthesizing 79 independent samples across 67 studies, racial disparities in juvenile justice processing persist net of legal factors (offense, prior record), with the magnitude varying across decision points — the quantitative synthesis of four decades of DMC research showing that differential offending alone does not account for the observed disparities.
Bearing on this claim: Meta-analysis of 79 samples/67 studies: racial disparities persist net of legal factors, varying by decision point.
doi.org/10.1080/07418825.2021.1915364

Supporting

Precedent Bridge

Mitchell is the doctrinal ceiling on this claim. The court accepted that the defendant had produced statistics on disproportionate treatment of Black juveniles nationally, in Colorado, and in the charging district — and still held that standalone statistical evidence, absent a showing that the compared individuals were similarly situated, seldom proves discriminatory effect and does not establish discriminatory purpose. A litigator citing this claim should pair it with individualized comparator evidence; the disparity science establishes the pattern, not the intent element.2026 CO 8

Limits & Counter-considerations

All designs are observational: unmeasured offense-severity or contextual differences could inflate estimates, while control variables contaminated by earlier system contact (prior record) can deflate them; jurisdictional heterogeneity is high, so pooled estimates mask local variation; self-reported offending has its own biases; Latino, Indigenous, and Asian disparities are thinner literatures; and evaluations of DMC-mandate reforms show limited success, complicating remedy inferences.