JPRC

Q&A Article with the JPRC Violence Prevention team: Lori Toscano, Shaun Ali & Kerwin Henderson

Violence Prevention

Q&A Article with the JPRC Violence Prevention team: Lori Toscano, Shaun Ali & Kerwin Henderson

by

Despite welcome declines in the homicide rate across the United States, estimates indicate that more than 18,000 persons died by homicide in 2023. The pain and suffering caused by such violence is dramatic, extending beyond direct victims to exert harmful impacts on family members and loved ones, first responders, and communities. Using data-informed strategies is essential for understanding the risk factors contributing to these tragedies, enabling interest-holders to develop targeted interventions that can disrupt these cycles of violence. But what tools are needed to improve this process?

We invited JPRC’s leaders in violence prevention to respond to this and other related questions. Lori Toscano is Director of Justice Technical Assistance, and Shaun Ali and Kerwin Henderson are Technical Assistance Specialists. They oversee several projects focused on helping local and school communities address violence and are the main technical assistance providers to local communities receiving violence prevention grants in Pennsylvania. They recently collaborated with our partner, Muflehun, to develop a new tool known as the Violence Prevention Navigation Framework (VPNF) to help communities address violence. The following answers represent collective responses from the team.

Violence prevention is a priority in the United States. However, you say there are gaps in how violence is addressed. What are the gaps?

Historically, violence prevention efforts have been fragmented and underfunded, forcing localities to focus primarily on short-term solutions. States and localities often struggle taking inventory of what violence prevention efforts already occur in their jurisdictions, leading to duplication of effort. Lack of data, specifically cross-sector data, can lead to large investments in siloed solutions and ineffective, often controversial, technology and strategies, such as surveillance and heavy-handed enforcement.

You have developed the VPNF process to help communities address these gaps. What is VPNF?

The VPNF is a new method for addressing gun violence. It uses big data, machine learning, and local partners’ collective knowledge to analyze community- and society-level risk and protective factors related to gun violence. Using that information, the VPNF process then prioritizes the factors that have the greatest impact, whether positive or negative, on local gun violence. With that list of factors, we then provide decision-makers with evidence-based recommendations for program and strategy investments that can have the biggest impact.

Let’s say a city wants to use VPNF. How does it help them?

VPNF provides state and local leaders with a clear understanding of the specific challenges that their communities face regarding gun violence. For a city, it gives them access to the full landscape of the quantitative data analysis which consists of more than 70 community- and society-level factors that act as risk factors for—or protective factors against—community gun violence. Our technology then analyzes these data and lists the factors according to their level of impact. It can also show how each factor has changed over time.

With this list, we have a starting point to dig deeper. Through interviews and surveys with relevant community and government stakeholders, including those directly impacted by the violence, we can work to identify why the factors are showing up as they are and can glean the people perspective to supplement the quantitative data. In the end, the city’s leaders have a clear picture of the community and society-level root causes for gun violence in their locality. We can then provide a series of recommendations of programs, policies, and strategies to address these identified causes, which can result in data-informed funding and strategic decision-making to meaningfully address root causes and historic inequities. VPNF provides a city with a stronger foundation for long-term gun violence prevention, based on comprehensive data and community input.

Once you have provided the data and recommendations to a jurisdiction, what happens then?

They will then need to work toward implementation, which can happen in a variety of ways. For instance, a jurisdiction may decide to use our recommendations to develop or revise their transit plan, reconsider locations of physical and mental health services, increase civic engagement opportunities, or pursue some combination of those approaches. Additionally, we can work with the jurisdiction on how to use the data and recommendations to bolster current and future violence prevention and intervention efforts. They may decide to implement initiatives to address the factors identified in our analyses, separately or as a part of a broader strategy or campaign. In any case, our team can provide technical assistance, consulting, and advisory services to help jurisdictions implement our recommendations.

There are many types of violence (e.g., targeted violence, mass shootings, gang violence). Can VPNF inform strategies for addressing all of them?

Yes, VPNF can be adapted to address other types of violence if risk and protective factors have been identified and robust data are available. In the same way that we are analyzing over 70 factors that contribute to or mitigate gun violence with VPNF, we can adapt the machine learning to analyze risk and protective factors for any other form of violence. The process for incorporating qualitative data would be similar, and so would the process for generating recommendations. Our hope is to learn from each VPNF engagement, initially focused on community-level prevention of gun violence, and to use those experiences collectively to evaluate, enhance, and adjust the process. Over time, we hope to use the lessons learned to adapt VPNF to additional forms of violence.

If communities or jurisdictions are interested in VPNF, what should they do?

For those interested in our framework and wanting to learn more about how VPNF can help them comprehensively address gun violence in their communities, they should visit the VPNF web page and complete the Contact Form. We will then set up a call to talk through their needs and how the JPRC can help.

What do we learn from data on homicides of school-aged youth?

by

The HEDCO Institute at the University of Oregon published this article on October 3, 2024 that amplifies data reported by the U.S. Department of Education’s National Center for Education Statistics. The tragic multiple casualty shootings at schools such as the September 2024 massacre at a high school in Georgia, creates anxiety over the safety of children and staff at school. In this article, the authors delve into data on homicides of school-aged children and report results that some find counterintuitive to the news cycle. Homicides of children at school (including afterschool and sporting events) represent a small fraction (1-2%) of total youth homicides; most youths are victims of homicide outside of schools (i.e., on the streets, in their homes, and at other venues). These findings prompt a call for evidence-based community-based and school-based strategies addressing youth violence broadly.

A revised version of this article, entitled Look for solutions beyond school grounds to address youth homicides, was published on October 28, 2024 by EdSource. Published since 1977, EdSource is the leading independent news source on California education.

NIJ-funded study on safe firearms storage

by

We are partnering with the Texas School Safety Center at Texas State University on a new NIJ-funded grant. The grant supports a 3-year study to examine early-stage implementation of a recent Texas mandate requiring schools to disseminate safe firearms storage information to students’ families each year. The research team will survey a statewide representative sample of school administrators and parents to assess perceptions, attitudes, and behaviors related to the campaign, including how the findings vary across demographic and community factors. The study will also include a process of co-interpretation with students and parents to inform future implementation strategies in the state.

Intensified Support for Juvenile Offenders on Probation: Evidence From Germany (RDD Design in Crime and Justice Project Study Summary #2)

by

WestEd’s Jonathan Nakamoto and Trent Baskerville summarized this study conducted by researchers Christoph Engel, Sebastian J. Goerg, and Christian Traxler to investigate the impact of an intensive probation program on youth recidivism.

Using a prospective Regression Discontinuity Design (RDD), the study assigned half of the eligible youth above a cutoff score to receive the intensive probation program, while the other half received regular probation. The research team collected individual-level conviction data from Germany’s Federal Crime Register and used a risk scorecard to evaluate each youth’s criminal history and risks/needs.

The findings revealed that recidivism was approximately 10 percentage points lower for youth eligible for the intensive probation program after 6 months, with this difference increasing to around 30 percentage points between 1 and 3 years later. These reductions in recidivism were statistically significant for less severe offenders close to the assignment variable cutoff. Overall, the study highlights how intensified personal support and increased monitoring by probation officers can contribute to positive outcomes in reducing recidivism among juvenile offenders.

Under the Cover of Darkness: How Ambient Light Influences Criminal Activity (RDD in Crime and Justice Project Study Summary #3)

by

WestEd’s Jonathan Nakamoto and Trent Baskerville summarized this study conducted by researchers Jennifer L. Doleac and Nicholas J. Sanders to investigate the impact of increased daylight during evening hours due to Daylight Savings Time (DST) on criminal activity.

Using a retrospective Regression Discontinuity Design (RDD) study, the researchers used crime and population data from 558 jurisdictions in the United States between 2005 and 2008. They found that DST led to reductions in robbery and rape rates, with the most significant decreases occurring in the hour following sunset.

The study revealed statistically significant reductions of 7 percent for robbery and 11 percent for rape in daily crime rates after DST, while there were no significant changes for aggravated assault and murder. However, analysis of crime rates specifically within the hour after sunset revealed larger reductions of 27 percent for robbery and 38 percent for rape. Importantly, the findings indicated that criminal activity was not shifted to other times of the day. The authors estimated that these reductions could result in annual savings of nearly $60 million due to avoided robberies and approximately $250 million due to avoided rapes as a result of DST implementation.

Examining Prison Effects on Recidivism: A Regression Discontinuity Approach (RDD in Crime and Justice Project Study Summary #1)

by

This study summary prepared by WestEd’s Jonathan Nakamoto and Trent Baskerville explores the impact of prison sentences on recidivism rates. The study, conducted by Ojmarrh Mitchell, Joshua C. Cochran, Daniel P. Mears, and William D. Bales, focuses on Florida’s point system for sentencing and uses a Regression Discontinuity Design (RDD) to evaluate the effects of prison versus nonprison sanctions on subsequent reconviction rates.

The authors analyzed data from the Florida Department of Corrections and found that there was no statistically significant difference in reconviction rates between cases just above and cases just below the cutoff score for imprisonment. This suggests that prison sentences did not lead to a reduction in recidivism over a 3-year period for mostly nonviolent property and drug offenders near the cutoff score. The study concludes that alternative sanctions may be more effective for these types of offenders considering he high costs associated with imprisonment.