If we want Community Violence Intervention (CVI) to be seen—and funded—as essential public safety infrastructure, we must continue building the evidence base to demonstrate its impact. But we also must expand how we define and measure success.
CVI saves and changes lives. We’ve seen it in the neighborhoods where violence has dropped, where once-disconnected individuals are now leading change, and where entire communities have reclaimed hope, and greater purpose. But too often, traditional research tools and academic norms struggle to capture these outcomes in ways that resonate with policymakers, funders, and the public.
The future of CVI depends on an evolved approach to research—one that values lived experience, centers equity, and is designed not just to assess, but to elevate what works.
Most CVI programs are expected to prove their value through reductions in shootings and homicides. Indeed, these are critical metrics. But they are also lagging indicators—they show us what happened after violence occurred or was prevented.
What gets missed in this narrow frame are the leading indicators:
- The conflicts de-escalated before a shot was fired.
- The cycles of retaliation interrupted.
- The lives stabilized through intensive mentoring, access to healing, life skills training, and leadership development opportunities.
- The trust earned in communities where trust is hard to come by.
When CVI is working, it often looks like “nothing happened.” But that nothing is everything. To get to ‘nothing happened’ requires a persistent relentlessness in the trenches of conflict, anger, fear, and pain, every single day. As Advance Peace national strategy specialist Freddie Dearborn likes to say, “we must be willing to work 25/8/366.” This is CVI.
Traditional research methods—particularly randomized controlled trials—can be ill-suited for CVI. Why? Because CVI isn’t a clinical intervention delivered in a controlled setting. It’s a dynamic, relational, and adaptive practice embedded in the real lives of real people facing real danger.
What’s needed is a broader understanding of what counts as credible evidence. That includes:
- Mixed-Methods Approaches
Combine quantitative data (e.g., shooting reductions, service engagement, employment and education outcomes) with qualitative insights from frontline practitioners, fellows, and community members. - Community-Led Evaluation
Partner with those closest to the work to define success, co-design evaluation tools, and interpret findings. This improves both relevance and accuracy. - Practice-Based Evidence
CVI is grounded in practitioner wisdom. The field should elevate this experiential knowledge as a valid and essential form of evidence—not as anecdote, but as an accumulated, tested strategy. - Process Evaluation
Understand how CVI works. What relationships, rituals, training, and systems are in place that led to optimal impact? Process evaluation ensures CVI efforts are replicable without losing their center.
To support this expanded research agenda, we need a stronger infrastructure:
- Universities, public agencies, and CVI organizations must work as equal partners—not as subjects and observers, but as co-producers of knowledge.
- Short-term funding yields short-term analysis. To understand CVI’s full impact, we need longitudinal studies that track individuals and neighborhoods over years—not months.
- CVI data must be handled ethically, with safeguards that prevent surveillance and promote healing. Data sharing should empower communities, not expose them.
- Research findings should be translated into tools, trainings, and policies that help the field improve—not just satisfy external requirements.
Innovative research efforts are already expanding our understanding of CVI:
- Studies of the Richmond Office of Neighborhood Safety Advance have shown promising reductions in gun violence, while also documenting improved life outcomes for those it serves.
- Local evaluations in cities like Chicago, New York, and Oakland are demonstrating how community-based strategies can complement and outperform traditional enforcement in specific violence-prone areas.
- National initiatives—such as the DOJ’s community violence intervention and prevention initiative (CVIPI) grants and the White House Office of Gun Violence Prevention—have helped us get closer to standardized data collection and support for more rigorous evaluation.
But the research ecosystem remains fragmented. Too many efforts operate in isolation, and too many evaluations are disconnected from practice. We need to unify the field—not around a single model, but around a shared commitment to learning and accountability.
Researchers must meet the field where it is, not where academia traditionally operates. And funders must resource evaluation in ways that don’t burden the very efforts they’re trying to support.
CVI is building peace in places long abandoned by systems. If that work is to continue and grow, we need a research movement that is just as bold, just as relational, and just as rooted in the realities of the people most affected by violence.
Making CVI permanent means building a body of evidence that speaks not just to what we’ve stopped—but to what we’ve started.