OSINT · Publicly Available Sources INVESTIGATION ORACLE

Thirty Hours and No License: How Untrained CASA Volunteers Steer America's Family-Separation Machine

July 26, 2026 OPUS · Claude Opus Project Milk Carton SSI PI License #5337

Thirty Hours and No License: How Untrained CASA Volunteers Steer America's Family-Separation Machine

Every year, tens of thousands of American children have the most consequential decision of their lives — whether they go home, get adopted by strangers, or linger for years in foster care — shaped by the written recommendation of a volunteer who completed roughly 30 hours of training, holds no professional license, is bound by no rules of evidence, and can never be sued for getting it wrong. This is the Court Appointed Special Advocate (CASA) and volunteer guardian ad litem (GAL) system, a federally seeded network that judges lean on so heavily that a CASA report often becomes the judicial finding verbatim. It runs on roughly $12–15 million a year in federal appropriations funneled through the Department of Justice, is staffed by a volunteer corps that is roughly 90 percent white speaking for a foster population that is disproportionately Black and Native, and — after nearly half a century and hundreds of millions of taxpayer dollars — has never been shown by rigorous evidence to help children. The best national evaluation ever commissioned found the opposite: children assigned a CASA were more likely to be removed from home and less likely to be reunified with their families. The organization at the top spent much of the last three years as a federal "high-risk" grantee that had to repay taxpayer money for "unsupported, unallowable" spending. This is that story.

How the Machine Actually Works

CASA began in 1977 in the Seattle courtroom of Superior Court Judge David Soukup, who wanted lay volunteers to gather background on the children whose fates he was deciding. The model spread, and in 1990 Congress wrote it into federal law through the Victims of Child Abuse Act (codified at 34 U.S.C. § 20301 et seq.), authorizing the Office of Juvenile Justice and Delinquency Prevention (OJJDP) inside the Justice Department to fund the expansion of CASA programs nationwide. Today the National Court Appointed Special Advocate/Guardian ad Litem Association for Children — a Seattle-based 501(c)(3), EIN 91-1255818 — sits atop a network of roughly 900 local programs staffed by tens of thousands of volunteers serving over 200,000 children a year.

Here is the mechanism that matters. In an abuse or neglect (dependency) case, a judge appoints a CASA volunteer — in many states the volunteer is the guardian ad litem, the person legally charged with representing the child's "best interests." The volunteer visits the child, talks to parents, teachers, caseworkers, and foster parents, and writes a report to the court recommending what should happen: whether the child should be removed, whether reunification should be pursued or abandoned, whether parental rights should be terminated. National standards require at least 30 hours of pre-service training and 12 hours of continuing education a year. That is the entire credential. No degree in social work, psychology, child development, or law is required. No licensure. No board oversight. No malpractice exposure.

Then the recommendation lands on a judge's desk in a courtroom running on an assembly-line docket, where hearings are frequently measured in minutes and the CASA is often the only person who claims to have spent real time with the child. Judges, structurally starved of independent information, adopt CASA recommendations at high rates — and because the volunteer is cloaked in the moral authority of the selfless citizen "speaking for the child," those recommendations are rarely challenged. As a Texas Law Review analysis ("Procedural Due Process and the Improper Expert Opinions of Court-Appointed Special Advocates") documents, CASA reports routinely contain exactly the kind of opinion testimony — psychological assessments, predictions of future risk, judgments about a parent's fitness — that would require a qualified expert under Federal Rule of Evidence 702 and the Daubert standard if a witness offered it from the stand. Yet "CASA reports and testimony are rarely challenged on evidentiary grounds." The volunteer's untested, unsworn, hearsay-laden narrative slides into the record and becomes fact. A parent facing the permanent loss of a child confronts an accuser who never had to qualify as an expert, whose sources need not be disclosed, and whose conclusions face no Daubert gate.

The Money: A Federal Program That Never Proved It Works

The federal footprint is modest in dollars but decisive in leverage. Congress funded the CASA program at $12 million in fiscal years 2018, 2019, and 2020, bumped it to $12.5 million in FY2021, and has twice pushed it to $15 million; in FY2024 the program distributed nearly $13 million in subawards for program growth, volunteer recruitment, strategic-planning demonstrations, and capacity building. That federal money flows DOJ → OJJDP → National CASA/GAL, which then re-grants it as subawards to state and local programs — the classic pass-through structure that makes the national organization the indispensable gatekeeper and the local programs financially dependent on it.

What the federal government has never bought with that money is evidence that the program works. This is the central scandal hiding in plain sight. A 2007 federal audit found National CASA had no outcome-based performance measures — meaning the government was funding a child-welfare intervention without measuring whether children were safer, reunified faster, or better off. Nearly two decades later, that gap remains: there is no Government Accountability Office effectiveness audit of CASA, and the peer-reviewed literature, as researchers repeatedly note, is "scant" and produces "mixed results" that "do not provide strong empirical evidence" of improved safety, permanency, or wellbeing.

Worse than absent evidence is the evidence that exists. The most ambitious national evaluation ever commissioned — the Caliber Associates study prepared for National CASA itself (2004) — found that children assigned a CASA volunteer were more likely to be placed out of the home and less likely to be reunified with their families or placed with kin. In the reading of the National Coalition for Child Protection Reform, the study's real "accomplishments" were to prolong the time children languished in foster care and to reduce the chance a child would be placed with relatives. Later audit work found CASA-involved children spent on average 1.2 to 2 months longer in foster care. Defenders offer a genuine methodological caveat — CASAs are disproportionately assigned to the most serious maltreatment cases, so the comparison isn't clean — and a 2020 study in Child Welfare by Cynthia Osborne and colleagues tried to isolate the effect with better controls. But the honest bottom line is damning for a program of this age and reach: after 48 years and hundreds of millions of dollars, its own flagship evaluation pointed toward more separation and less family reunification, and no rigorous study has ever cleanly demonstrated benefit.

The Named Player at the Top — and Its Accounting Problems

The National CASA/GAL Association is not merely an idealistic charity; it is a federal grantee that spent a remarkable stretch of the recent past under financial cloud. According to reporting by The Imprint, a federal financial review questioned more than $2.7 million in the organization's expenditures and found "inadequate" policies and procedures for "accounting, payroll, subrecipient monitoring, and progress reporting," together with high turnover among chief financial officers. In a March 2023 letter, the Justice Department laid out those deficiencies; the feds froze the group's funding and designated it a "high-risk" grantee — the government's formal label for an organization it does not trust to manage taxpayer money without special conditions. National CASA ultimately repaid more than a quarter-million dollars in "unsupported, unallowable" spending and remained under high-risk status for roughly 19 months before DOJ lifted the freeze and renewed funding.

So the incentive structure is worth stating plainly. The national organization's revenue and institutional survival depend on the federal appropriation and on maintaining the network of local programs it re-grants to. Its public messaging is relentlessly self-promotional ("Change a Child's Story"), while its own commissioned evaluation and the federal audits it has weathered tell a far more troubling story. Local programs, dependent on national subawards and on maintaining warm relationships with the judges who appoint their volunteers, have every reason to keep producing recommendations that please the bench and to avoid the kind of adversarial testing that might reveal how thin the underlying expertise is. Nobody in this chain — not the national office, not the local program, not the individual volunteer — has a financial or professional incentive to audit outcomes, because outcome data is the one thing that could threaten the money.

The Race Problem: A White Volunteer Corps Deciding the Fate of Black and Native Children

The demographic mismatch at the heart of CASA is not a footnote — it is arguably the system's defining feature. Volunteers are overwhelmingly white and female: studies place the corps at roughly 90 percent white (8 percent African-American) and 79 to 91 percent female. The children they speak for are disproportionately children of color — one dataset showed a served-child population that was 48 percent white and 36 percent Black or biracial, in a foster system where Black and Native American children are removed at rates far exceeding their share of the population.

The problem is not merely optics; it shows up in behavior. Research found volunteers spent a mean of 2.67 hours on African American children's cases versus 4.30 hours on the cases of children of other races — less attention, documented in the numbers, for the very children most at risk of separation. Legal scholars Amy Mulzer and Tara Urs, in their influential article "However Kindly Intentioned: Structural Racism and Volunteer CASA Programs," argue that the CASA model is almost designed to launder bias: it invites well-meaning, mostly affluent white volunteers to walk into poor Black and Native households and render subjective judgments about what a "good home" looks like, with no objective criteria, no evidentiary discipline, and the full authority of the court behind them. A middle-class volunteer may read a bare refrigerator, a shared bed, or a distrustful parent as neglect or danger when it is poverty, culture, or the entirely rational fear of a family that has learned the system takes children. Because CASAs "opine on what they think is best for children without any objective criteria," as critics put it, the discretion becomes the hiding place for racism — the mechanism The Imprint summarized as "with child welfare, racism is hiding in the discretion."

There is a proven alternative that exposes how much of this is bias rather than fact-finding. In Nassau County, New York, officials experimented with "blind removal" meetings — stripping the family's race and neighborhood from the case information presented to decision-makers. The result was a significant reduction in the removal of Black children. If simply hiding a family's race changes the removal decision, then race was driving the decision. The CASA model does the opposite of blind removal: it inserts an untrained, largely white, unaccountable human between the family and the judge and asks that person to render exactly the kind of subjective "best interests" judgment most vulnerable to bias. The movement that briefly trended as #CASASoWhite was pointing at precisely this.

The Accountability Gap: Immunity, No Evidence Rules, No Audit

Ask who is supposed to be watching, and the answer is almost no one — by design at nearly every level.

In court, the evidentiary safeguards that constrain every other participant do not bind the CASA. Prosecutors face rules of evidence; expert witnesses face Daubert; caseworkers face cross-examination and agency policy. The volunteer's report, as the Texas Law Review documented, sails past FRE 702 because nobody objects, importing hearsay and lay psychological opinion into findings that sever families.

In the civil courts, the volunteer and the GAL are shielded from consequences. Guardians ad litem enjoy absolute or quasi-judicial immunity in state after state — Ohio, Illinois, and Michigan among them — for conduct within their appointed role. The American Bar Association Journal has reported appellate rulings granting GALs absolute immunity even against allegations of wrongful conduct; in one documented case a GAL was accused of blocking the reporting of child abuse, refusing to allow a child's therapist to testify, and telling a mother she would lose custody if she didn't settle — and immunity still shielded most of the conduct. A narrow exception exists (the Pennsylvania Supreme Court declined to extend quasi-judicial immunity to a GAL facing a legal-malpractice claim), but the national rule is stark: the one actor whose recommendation can permanently end a parent-child relationship is the one actor a wronged family generally cannot sue. Licensed social workers can lose their licenses; lawyers can be disbarred; CASAs, holding neither, face neither.

At the federal level, the oversight that does exist has been financial, not substantive. DOJ audited the money and found it wanting — the high-risk designation, the repayments, the "inadequate" accounting. But no federal body has ever conducted the rigorous effectiveness audit that would ask the only question that matters for children: does a CASA appointment make a child safer, or reunify a family faster, or reduce time in care? The 2007 finding of no outcome measures was never truly cured.

The most recent chapter exposed how politically contingent even the funding oversight is. On April 22, 2025, DOJ issued more than 360 grant-termination notices, and National CASA's three OJJDP grants were among them — terminated, the notices said, because they "no longer effectuate the program goals or agency priorities." Two days later the organization suspended all OJJDP-funded services and subaward disbursements to the network, an abrupt cutoff that local programs said threatened services to more than 200,000 children. Then, within weeks, the funding was restored — but only after National CASA scrubbed diversity language from its own training materials. Beginning May 15, 2025, CASA leaders instructed trainers to "Skip or Adapt as Directed" dozens of sentences across slide decks, the volunteer manual, handouts, and the facilitator guide, to comply with the administration's directives; The Imprint reported the grants were restored after the removal of DEI terms. Attorney General Pam Bondi's July 29, 2025 guidance made explicit that phrases like "cultural competence," "lived experience," and "geographic targeting" would be treated as unlawful proxies for protected characteristics. So the very training content meant to counteract the program's documented racial-bias problem — cultural competency instruction — became the price of keeping the federal check. An organization already 90 percent white was pressured to teach less about race, not more.

Why It Matters, and What Would Actually Fix It

None of this is abstract. Every one of these recommendations is attached to a child and a family. When a 30-hour volunteer with no license writes that a mother is "not making progress" and a judge adopts it, a parent can lose a child forever. When volunteers spend fewer hours on Black children's cases, that disparity converts into removals and terminations. When the only national evaluation shows CASA-linked children are less likely to go home and more likely to be taken from kin, the "help" is doing measurable harm to the constitutional interest — a parent's right to raise their own child — that the Supreme Court has called among the oldest fundamental liberty interests. And foster care is not a safe harbor: children who age out of it face grim outcomes in homelessness, incarceration, and trafficking. A system that keeps children in care longer and reunites them less is not a neutral disappointment; it is an active driver of the harm.

The fixes are not mysterious. Require outcome auditing — condition every federal dollar on rigorous, independent, race-disaggregated measurement of whether CASA involvement improves safety, permanency, and reunification; if it can't be shown to help, stop funding it. Impose evidentiary discipline — CASA reports offering opinion testimony should meet the same reliability standards as any expert, and defense counsel should be trained and expected to challenge them under Daubert. Strip or narrow immunity for grossly negligent or reckless conduct, so the actor with life-altering power bears some accountability. Professionalize or replace the role — many reformers argue the volunteer-GAL model should give way to trained, licensed, well-resourced attorneys and social workers who actually represent the child, or to the child's own expressed wishes, rather than a stranger's subjective "best interests" guess. Adopt blind-removal-style safeguards that keep race and neighborhood out of the recommendation. And fund families first — kinship placement, anti-poverty support, and reunification services — instead of paying a middle layer of untrained advocates who, the data suggest, tilt the scale toward separation.

The CASA system was built on a genuinely humane impulse: that a child in a cold courtroom deserves someone in their corner. But good intentions were never audited, never licensed, never made accountable, and never proven to work — and for the disproportionately poor, Black, and Native families ground through dependency court, "kindly intentioned" has too often meant a stranger with 30 hours of training and lifetime immunity holding the pen that decides whether their children ever come home.