Schools Are Spending Billions on Safety. Nobody Can Prove It's Working.

A California family sues their district after their daughter faces sustained, documented bullying over two academic years. The district had an anti-bullying program. Staff received training. There were posters. There was a policy binder. The bullying continued anyway, and now attorneys are involved, and the district's legal team is doing what legal teams do: scrambling.

The family's question was simple. The district could not answer it: What evidence do you have that your program actually stops bullying?

That question is eating school administrators alive right now, and most of them will not say so publicly.

Part One: The Pressure

Schools are being squeezed from every direction simultaneously, and the financial stakes are not abstract.

Federal civil rights enforcement offices have put districts on notice that failure to address peer harassment, particularly harassment targeting protected characteristics, can trigger federal investigations and the loss of federal funding. State legislatures have passed wave after wave of school safety mandates following high-profile incidents, with compliance costs that land directly on district budgets regardless of whether those mandates come with funding attached.

Parent advocacy groups, armed with attorneys and social media reach, have made it genuinely risky for a school board to say "we're not sure this program works" at a public meeting. So boards approve the programs. The checks get written.

The American school safety industry, including hardware, software, mental health staffing, training programs, and curriculum products, has grown into a multi-billion-dollar market. Individual districts have committed millions to threat assessment teams, anonymous tip lines, social-emotional learning curricula, restorative justice coordinators, and security infrastructure, sometimes all at once, sometimes without a coherent theory of how these pieces interact.

Legal liability drives the logic. If a district can show it spent money, hired staff, and ran programs, it has a defense. Whether those programs reduced harm is a separate question, and for legal purposes, often a secondary one.

Part Two: The Blind Spot

Here is the sentence that district safety officers and superintendents will say in private but rarely in board meetings: We don't have the outcome data to know if this is working.

That admission is more common than it should be. Districts typically track inputs: dollars spent, staff hired, programs implemented, trainings completed. They track incidents reported, which is its own problem because better reporting systems can make a safer school look more dangerous on paper. What most districts do not track rigorously is whether specific interventions caused measurable reductions in specific harms.

The research base for many popular safety programs is genuinely thin. Some widely-adopted bullying prevention curricula have been evaluated in controlled studies and shown modest effects at best, effects that shrink further when programs are implemented by non-specialist staff under real school conditions rather than research conditions. Some threat assessment models have solid peer-reviewed backing. Many do not.

Districts rarely have the internal capacity to evaluate what they are buying. They rely on vendor-provided data, which has obvious conflicts of interest, or on program reputations that spread through administrator networks rather than through evidence review. The pattern will look familiar to anyone who followed the reading intervention evidence gap: widespread adoption of programs that sounded rigorous, combined with limited independent verification that they actually worked at scale.

One factor compounds the problem: school safety is politically charged enough that admitting a program failed feels dangerous. If a board acknowledged that its $400,000 SEL curriculum produced no measurable change in disciplinary incidents, the next news cycle writes itself. So the program continues. A new program gets added on top of it.

Part Three: What Actually Works

The evidence base is not empty. It is just less exciting than the sales pitch.

Threat assessment programs with structured protocols have the strongest research support for preventing targeted violence. The model developed through research with the Secret Service and later expanded through university-based research centers focuses on identifying students on a pathway toward violence and connecting them with intervention, rather than on profiling or punishment. Districts that implement this with fidelity, meaning trained teams, consistent case review, and documented follow-through, show real results. The challenge is fidelity: the model is relationship-intensive and requires administrator buy-in that not every district sustains.

Cognitive-behavioral intervention for students already showing elevated risk has strong randomized trial evidence behind it. This is not universal SEL curriculum delivered to an entire grade level; it is targeted, skill-based work with students who are already struggling. The problem is it requires licensed clinicians, which requires money and staff retention, which is where many districts stall.

Restorative practices, when implemented as a genuine shift in school culture rather than as an add-on to existing punitive systems, show promising evidence for reducing repeat disciplinary incidents and improving school climate. The research quality varies by study design, but the better-designed evaluations are encouraging. The failure mode is districts that add a restorative coordinator without changing how teachers and administrators respond to conflict daily, which produces the coordinator's caseload but not the culture shift.

Adult connectedness and mentoring has consistent backing across multiple research reviews. Students who report having at least one trusted adult at school show lower rates of a wide range of risk behaviors. This is cheap to implement and hard to scale because it depends on adult behavior change, not a purchased program.

Notably, the interventions with the best evidence tend to be relationship-based, staff-intensive, and hard to purchase as a packaged product. The interventions that are easiest to purchase and demonstrate to a school board tend to have weaker or mixed evidence. That gap is not a coincidence.

Adoption of evidence-based programs is slower than it should be for predictable reasons. Evidence-based programs often require more staff training time, more administrative oversight, and more patience before results appear. They are harder to present at a board meeting than a new security camera system or a branded curriculum with a glossy implementation guide. And the political pressure on schools is immediate, while evidence takes years to accumulate.

This dynamic is not unique to safety. Parents watching male student disengagement grow into a workforce crisis will recognize the shape of it: a real problem, a flood of interventions, insufficient accountability for whether any of it actually helps.

What Parents Should Actually Ask

If your student is in a district spending significantly on safety programs, these questions will tell you whether the district is operating from evidence or from liability management.

• What specific outcomes does this program measure, and what does your data show? Not participation numbers. Not training hours. Outcome data.
• What independent evaluation supports this program? Vendor-provided testimonials are not evaluation. Ask for peer-reviewed studies or independent program assessments.
• How does the district evaluate whether a program is working after it's implemented? If there is no answer, the district is not evaluating.
• What did you stop doing or change based on evidence it wasn't working? A district that has never discontinued a program has never evaluated one.
• Who has access to the outcome data, and is it reported publicly?

The red flags are straightforward: significant budget commitments, no outcome metrics, no independent evaluation, and leadership that responds to these questions with descriptions of inputs rather than results.

Families who are already outside the traditional district model, including those using online options like Fresno Online High School or Round Rock Online High School, sometimes choose that path precisely because the traditional environment felt unsafe and the district's response felt like theater. That is a legitimate calculation. But even in alternative settings, the right question remains the same: what does the evidence show, and who is measuring it?

Schools are not villains in this story. They are operating under genuine legal exposure, enormous political pressure, and inadequate research infrastructure, and they are making spending decisions under those conditions. That context matters. It does not, however, change what parents are owed: a straight answer about whether the money being spent on their student's safety is actually keeping their student safer.

If the district cannot answer that question, the question needs to get louder.