Thirty Years
of Evidence.
One Mission.
312 peer-reviewed publications. 47 states reached. 14 model laws enacted. The research that shapes how America protects its most vulnerable children.
Three Decades.
Documented.
Every finding built on the last. Every dataset expanded the question. This is not a history — it is the accumulation of evidence that earned Advocate its seat at every table that matters.
Why do children re-enter foster care after reunification?
Dr. Eleanor Marsh convened twelve caseworkers in a church basement in Columbus, Ohio. The question on the table was deceptively simple. The answer took three decades to fully document. Early qualitative interviews with 240 families across four counties revealed that reunification failure was not random — it was systematically predicted by service gaps that agencies could measure and address.
Longitudinal tracking reveals a 7-year pattern invisible to annual reports.
The Ohio Permanency Cohort Study became the first dataset to track 1,800 children across seven years of child welfare contact. Where annual agency reports showed improving reunification rates, the longitudinal data revealed a troubling counter-pattern: 34% of reunified children re-entered care within 36 months. This finding reached the U.S. Senate Subcommittee on Children and Families in 2003.
The data reaches Capitol Hill during CFSR Round Two.
Invited to testify before the Child and Family Services Review panel, Advocate presented findings that directly shaped the federal outcome measures now used in all 50 states. The institute's recommendation to weight "stability of placement" as a primary outcome — rather than a secondary indicator — was adopted verbatim into the CFSR Program Improvement Plan framework.
The Kinship Stabilization Act passes in 12 states.
Drawing on a decade of kinship placement outcome data, Advocate drafted the Kinship Stabilization Act — model legislation providing tiered financial support and training access to relative caregivers. Between 2015 and 2022, 12 states enacted versions of the Act. A 2023 evaluation found a 28% reduction in kinship placement disruption in adopting states compared to a matched control group.
AI-assisted risk screening meets 30 years of ground-truth validation.
Advocate is now the only research body with a validated 30-year longitudinal dataset against which predictive risk screening algorithms can be tested for racial and geographic bias. Our current NIH-funded study is the first to assess whether algorithmic tools improve or replicate the disparities documented in our earliest qualitative interviews. The answer will determine how the next generation of child welfare technology is deployed.
The Ohio Permanency Study:
What Annual Reports Miss
When Ohio's annual child welfare report showed reunification rates improving year-over-year, the state celebrated. Our longitudinal tracking of the same 1,800 children told a different story: one in three was back in foster care within three years. The difference between those two numbers is the difference between a system that feels like it's working and one that actually does.
"The data doesn't lie — it waits. It waits until you follow it long enough to see what annual snapshots cannot."
Findings in your inbox,
before they reach the floor.
Six policy briefs per year. Each one built on primary data. Read by state directors, foundation officers, and Senate staff before markup sessions.
How We Know
What We Know
Every claim we make is traceable to a primary source. Every recommendation we publish is backed by a methodology that has survived peer review, federal scrutiny, and the test of real-world implementation.
Longitudinal Cohort Design
Children tracked across 7+ years of child welfare contact. Not snapshots — full trajectories.
Mixed-Method Integration
Qualitative field interviews coded and matched against administrative data at the case level.
Peer-Reviewed Publication
312 publications across Child Welfare, Social Service Review, and Journal of Public Child Welfare.
Practice-Embedded Research
Embedded researchers in 23 county agencies. Data collection happens where decisions are made.
You've Read the Evidence.
Now Access It.
312 publications. 30 years of primary data. Policy briefs written for the people in the room when decisions are made. Access requires only professional context — no paywall, no institutional subscription.
Kinship Stabilization at Year Eight
Outcome data from 12 adopting states, 2015–2023
Ohio Permanency Cohort Study, Vol. 4
Updated 2019 with 1,800-child dataset
CFSR Round Two: Federal Outcome Measures
U.S. Senate Subcommittee on Children and Families
Trauma-Informed Reunification Protocol
Adopted by 9 state agencies, 2020–2024
Who is this library for?
Request Library Access
Professional context only. No payment, no institutional subscription.