The U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention (OJJDP) funded Gang Reduction Program (GRP) was a targeted multiyear (2003–08) initiative created to reduce crime and violence associated with youth street gangs in a select group of cities across the United States. The initiative formed collaborations involving federal agencies, local stakeholders, and their communities to create a comprehensive, integrated and coordinated program. The program included primary prevention, secondary prevention, intervention, and gang suppression programming. The GRP was intended to address individual needs and risk as well as communitywide issues. For a visual depiction of the GRP Framework, please see Cahill and Hayeslip (2010, 3).
The GRP integrated the Spergel model of gang interventions (also known as the OJJDP Comprehensive Gang Model), parts of Project Safe Neighborhoods, and other OJJDP–funded programs to form one comprehensive approach to gang reduction. According to the Spergel model, gang problems result from the interaction of sociological, demographic, economic, and cultural factors along with social instability and lack of economic opportunity. The model concentrates on assessing the needs of youths and providing them with individualized support services and suppression/control by involving their families, local organizations, and their communities (Cahill et al. 2008)
GRP communities were smaller (2 to 5 square miles on average), had strong citizen involvement, and had substantial gang activity and crime. The final sites were selected at the discretion of OJJDP. The Southside community of Richmond, Virginia, was selected because of increasing conflict between local, “homegrown” African American gangs and national (Latin Kings) and international (Mara Salvatrucha aka MS–13) Hispanic gangs that appeared as the Hispanic immigrant population began to rise in the area. There was also concern that the gang problem was due to African American gangs and Hispanic gangs preying on Hispanic day laborers.
The Richmond GRP (known locally as the Gang Reduction and Intervention Program, or “GRIP”) focused efforts on eight domains: child welfare, community, crime, education, employment, health, physical environment, and probation/corrections. The goal of the Richmond GRP, as for OJJDP’s overarching GRP, was to reduce youth gang crime and violence through a combination of primary prevention, secondary prevention, intervention, and suppression strategies.
Richmond GRP leadership included a multidisciplinary intervention team and street outreach worker. The intervention team led by the Richmond GRP Program Coordinator consisted of representatives from probation, law enforcement, social services, schools, and other services providers. Together the team and outreach worker:
- Implemented outreach activities to promote public awareness
- Referred youths and families to GRP services
- Conducted risk assessments and developed intervention plans
- Provided case management
Using part of the funds from OJJDP, the Richmond Police Department (RPD) created a citywide gang unit and added additional car, bike, and foot patrols in the target areas. In addition to the typical suppression services of directed patrols, the RPD provided a comprehensive range of prevention and intervention activities to work with the community to address crime and other community issues. The RPD provided programs and services designed to increase police involvement with the program and familiarize residents with police procedures and activities. The department also provided truancy and dropout prevention programs to educate youths and their families about avoiding truancy and teach skills to promote positive behaviors.
Other services provided to the community by the Richmond GRP were:
- The One-Stop resource center, which identified at-risk youths before they became involved in gang activity and provided families access to services to strengthen the family bond and minimize the attractiveness of the gang lifestyle
- Mobile health care through Bon Secours Richmond Health Systems’ Care-A -Van services and prenatal and infant care provided by Obstetricians/Gynecologists in the area
- Reentry services that included housing, counseling, job training, life skills, residential treatment programs, intensive case management, and GED practice and training
- Tattoo removal services
- Mental health counseling, substance abuse counseling, and truancy and dropout prevention programs provided by Family Preservation Services
- Training for service providers and community members in Spanish to communicate with clients, as well as English as a Second Language classes for residents
For a complete list of programs implemented, see Cahill and colleagues (2008, Appendix G).
Cahill and colleagues (2008) used a quasi-experimental design to analyze the preliminary impact of the Office of Juvenile Justice and Delinquency Prevention’s Gang Reduction Program (GRP) on gang-related crime in Richmond, Va. At the time of the evaluation the suppression component of the Richmond GRP had likely exerted the most significant measurable effect on gang-related crime. The study areas included:
- The target area, which was selected because of its escalating level of violence and increasing gang-related conflicts. This area was the southern part of Richmond, west of the James River and south of downtown. The target area was 48 percent male, 67 percent African American, and 9 percent Hispanic. Nineteen percent of the target area population lived below the poverty level, and 6 percent of its households received public assistance.
- The comparison area, which was selected because of its demographic and criminogenic resemblance to the target area, was north of the James River. The comparison area was 47 percent male, 67 percent African American, and 2 percent Hispanic. Twenty-two percent of the comparison area population lived below the poverty level, and 3 percent of its households received public assistance.
- The displacement area, which was selected because of its spatial proximity to the target area and the availability of data, surrounded all but the southern portion of the target area. The displacement area was 45 percent male, 73 percent African American, and 4 percent Hispanic. Twenty-two percent of the displacement area population lived below the poverty level, and 6 percent of its households received public assistance.
Using police incident reports, researchers measured the volume of drug-related, serious violent, gang-related, and gang-related serious violent incidents in the study areas. Monthly time series were created for each study area from January 2002 through March 2007. Researchers considered intervention points from June 2005 through November 2005. Since some GRP resources were used to improve the quality and quantity of gang intelligence information, changes to outcome measures could have been due either to changes in number of gang incidents or changes in gang intelligence. To address this concern, researchers looked at serious violence in addition to gang-related crime. An interrupted time-series design was used to analyze results. The Auto Regressive Integrated Moving Average framework was used to estimate the effects of the intervention.
Using hospital admission data collected from Virginia Health Information, Inc., researchers analyzed admissions for the care of external injuries to assess changes in the level of gang violence. Measures included type of injury (gunshot, stabbing, fight/brawl, and other) and length of hospital stay (in days). Data was collected from 2004 to 2006, with the suppression efforts taking place midyear of 2005. Assessments were limited to pre–post descriptive analysis only, because of the lack of available data.
The evaluation team expected to collect data on the eight domains (child welfare, community, crime, education, employment, health, physical environment, and probation/corrections). However, because of the unavailability of data and the difference between planned and actual activities, researchers were able to collect information only on crime, health, and education. Researchers also noted that since a substantial part of the GRP initiative was devoted to prevention services, some effects would not be noticed until after the evaluation period.
In a follow-up to the 2008 Cahill and colleagues study, Hayeslip and Cahill (2009) analyzed the final impact of OJJDP’s Gang Reduction Program (Richmond, Va.). This follow-up included an additional year of monthly time-series data to assess the suppression component, extending the evaluation period from January 2002 through March 2008. Researchers used police records for reported crime incidents, arrests, and calls for service for all offense types within the city of Richmond. A structural break analysis was conducted.
For the education component, researchers collected information on GRP programs, which aimed to increase academic performance and commitment to education. School-level data was collected from the Virginia Department of Education for the 2004–05, 2005–06, 2006–07, and 2007–08 school years. Information from two schools in the target area (Thomas Boushall Middle School and George Wythe High School) was compared with control schools outside of the target area (Martin Luther King Jr. Middle School and Armstrong High School). Control schools were selected based on their similarity to the target school based on total enrollment, racial composition, and percentage of students eligible for the Free/Reduced Lunch program.
Measures included attendance rates, dropout rates, and rates of offenses per 100 students. Interpersonal offenses included offenses against students, offenses against staff, and offenses against all people. Behavioral offenses included inciting a riot, gang activity, trespassing, and disorderly conduct/disruptive behavior. Weapons-related offenses included the use, possession, or threat of bomb devices, firearms or toy guns, and knives/other weapons. Property-related offenses included arson, breaking and entering/burglary, theft, and vandalism. Total offenses included all of the above offenses plus alcohol, tobacco, and drug offenses, as well as all other offenses.
Although Cahill and colleagues (2008) found some significant findings, the preponderance of the preliminary results suggests that the suppression component of the Office of Juvenile Justice and Delinquency Prevention’s Gang Reduction Program (GRP) in Richmond, Va. had no effect on gang-related crimes. A follow up conducted by Hayeslip and Cahill (2009) supported these results.
Cahill and colleagues (2008) found that beginning in November 2005 the target area experienced a nonsignificant increase of 5.41 drug-related incidents a month; the comparison area saw a significant increase of 7.7 drug-related incidents a month during the same time. This suggests that the Richmond GRP had no significant effect in reducing or increasing the number of drug-related incidents a month in the target area. The displacement area saw a nonsignificant increase of 6.76 drug-related incidents a month starting in September 2005, suggesting that displacement was not a concern.
Serious Violent Incidents
Starting in July 2005, the target area saw a significant increase of 3.66 serious violent incidents a month. In contrast, the comparison areas, starting in August 2005, saw a significant decrease of 3.17 serious violent incidents a month. The displacement area also saw a significant decrease of 5.1 serious violent incidents a month, beginning in November 2005. The findings suggest that the Richmond GRP had a significant negative effect on the number of serious violent incidents in the target area, since the number of incidents a month increased in the target area but decreased in the comparison and displacement areas.
Starting in July 2005 the target area saw a significant increase of 15.7 gang-related incidents a month. The displacement area also saw a significant increase of 3.52 gang-related incidents a month. However, since the comparison group had too few incidents to analyze, there was not enough evidence to suggest the Richmond GRP had a significant negative effect on gang-related incidents.
Gang-Related Serious Violent Incidents
Starting in July 2005, the target area saw a nonsignificant increase of 3.55 gang-related serious violent incidents a month. The displacement area also saw a nonsignificant increase of 1.32 gang-related serious violent incidents a month, starting in November 2005. Again, since the comparison group had too few incidents to analyze, there was not enough evidence to suggest the Richmond GRP had any effect on gang-related serious violent incidents.
Types of Injury
There was a nonsignificant decrease in the proportion of patients admitted to the hospital for gunshot injuries and fight or brawl injuries in 2006 (a 10.7 percent and a 3.3 percent decrease, respectively). At the same time, however, there was a nonsignificant increase in the proportion of patients admitted to the hospital for stabbing injuries and other injuries (a 1 percent and a 13 percent increase, respectively). This suggests the Richmond GRP had no significant negative and no significant positive effect on the types of injuries requiring hospitalization.
Length of Hospital Stay
In 2006 there was a nonsignificant decrease of 1.8 days for the average number of days patients were admitted for external injuries and a 1.0-day nonsignificant decrease in the median number of days patients were admitted. This suggests the Richmond GRP had no significant effect on reducing the length of hospital stay.
Hayeslip and Cahill (2009) reported that starting in August 2007 the target area for OJJDP’s Gang Reduction Program (Richmond, Va.) saw a nonsignificant decrease of 8.5 drug-related incidents a month. The comparison and displacement areas also did not see a significant impact on their drug-related offenses.
Serious Violent Incidents
Starting in August 2007, the comparison group saw a nonsignificant decrease of 5.0 serious violent incidents a month. The target and displacement areas also did not see a significant impact on their serious violent offenses.
There were no significant differences found for gang-related incidents between the target area, the comparison area, and the displacement area.
Gang-Related Serious Violent Incidents
There were no significant differences found for gang-related serious violent incidents in the target area and the displacement area. Additionally, the comparison area had too few incidents to analyze.
School Attendance Rate
No significant differences were found for school attendance rates between any of the target schools and the control schools.
School Graduation Rate
There were no significant differences found for school graduation rates between the target high school and the control high school.
School Dropout Rate
The target high school saw a nonsignificant decrease in dropout rates, from 7.06 in the 2004–05 school year to 6.34 in the 2007–08 school year. Conversely, the control high school saw a nonsignificant increase in dropout rates from 5.90 to 6.48 during the same period.
School Offending Rate
Although not significant, the target middle school saw an increase in rate of total offenses per 100 students from 232.6 in the 2004–05 school year to 522.4 in the 2006–07 school year, and then a decrease to 476.3 in the 2007–08 school year. The specific offense that seemed to contribute most to the increase in total offenses was the behavioral offense rate, which increased from 74.4 in the 2004–05 school year to 370.6 in the 2006–07 school year and then decreased to 343.1 in the 2007–08 school year. Conversely, the comparison middle school saw a nonsignificant decrease, from 273.2 in the 2004–05 school year to 200.6 in the 2006–07 school year, but the rate increased to 283.3 in the 2007–08 school year.
Although not significant, the target high school saw an increase in the total offense rate from 187.0 in the 2004–05 school year to 213.9 in the 2006–07 school year, and the a decrease to 273.4 in the 2007–08 school year. The comparison high school also saw a nonsignificant increase in total offense rate, from 126.1 in the 2004–05 school year to 196.1 in the 2006–07 school year, but it then decreased to 184.0 in the 2007–08 school year.