NT OF ME J T US U.S. Department of Justice R T A I P C E E D

B

O J C S Office of Justice Programs F A V M F O I N A C I J S R E BJ G O OJJ DP O F PR National Institute of Justice JUSTICE National Institute of Justice R e s e a r c h i n B r i e f Jeremy Travis, Director June 1999 Issues and Findings National Evaluation of Discussed in this Brief: The National Evaluation of Operation Weed and Seed, a strategy to con- Weed and Seed trol violent crime, drug trafficking, and drug-related crime in targeted by Terence Dunworth and Gregory Mills areas and to provide a safe environ- ment for residents to live, work, and raise their families. From the initial Operation Weed and Seed represents an include 200 sites nationwide. Until three grant sites in 1991, Weed and ambitious Federal, State, and local effort recently, cities were funded either as Seed has grown to include 200 sites to improve the quality of life in targeted demonstration sites, which generally re- nationwide. The Weed and Seed high crime areas of America’s cities. First ceived between $500,000 and $750,000 programs in eight sites—Hartford, Connecticut; Manatee and Sarasota launched by the U.S. Department of Jus- annually over 4 consecutive years, or offi- Counties, Florida; Shreveport, tice (DOJ) in 1991, the program attempts cially recognized sites, which received Louisiana; Las Vegas, Nevada; Ak- to control violent crime, drug trafficking, much smaller amounts ($35,000 in some ron, Ohio; Pittsburgh, Pennsylvania; and drug-related crime and to provide a cases). Most sites currently (1999) re- Salt Lake City, Utah; and Seattle, Washington—were selected for the safe environment for residents to live, ceive about $225,000 annually. national evaluation of their imple- work, and raise their families. mentation and measurable effects The Weed and Seed programs in eight on crime and public safety. These are, by themselves, conventional sites were selected for the national evalu- law enforcement goals. What makes the ation of their implementation and mea- Key issues: Weed and Seed strate- gically links concentrated and en- Weed and Seed concept distinctive and surable outcomes related to crime and hanced law enforcement efforts to innovative is the means by which these public safety: Hartford, Connecticut; identify, arrest, and prosecute vio- goals are achieved: Community-focused Manatee and Sarasota Counties, Florida; lent offenders, drug traffickers, and human services programs and neighbor- Shreveport, Louisiana; Las Vegas, other criminals operating in the target areas (weeding) and commu- hood improvement initiatives are Nevada; Akron, Ohio; Pittsburgh, nity policing with human services— strategically linked with intensified, Pennsylvania; Salt Lake City, Utah; and including afterschool, weekend, geographically targeted law enforcement Seattle, Washington. These sites were and summer youth activities; efforts by police and prosecutors. Weed selected by DOJ as examples of different adult literacy classes; and parental counseling—and neighborhood and Seed is essentially a coordination aspects of Weed and Seed. In each of revitalization efforts to prevent and strategy; funding is only one tool among them, the evaluation focused on one or deter further crime (seeding). The many to achieve the program’s objectives. two Weed and Seed target areas. (See ex- eight evaluation sites were selected (See “What is Weed and Seed?”) hibit 2.) Although each site had its own because they provide examples of different aspects of Weed and distinctive crime problems, they all Seed in application. In each site, Weed and Seed is administered by the shared high rates of violent crime related the evaluation focused on one or Executive Office for Weed and Seed to drug trafficking and drug use. Most two Weed and Seed target areas. within DOJ’s Office of Justice Programs. sites had serious gang-related crime Although each site had its own As exhibit 1 shows, from the initial three distinctive crime problems, they all problems. shared high rates of violent crime grant sites—Kansas City, Missouri; Tren- related to drug trafficking and drug ton, New Jersey; and Omaha, Nebraska— The effectiveness of weeding and seeding use. Most sites had serious gang- in 1991, Weed and Seed has grown to activities varied across the eight sites. related crime problems.

continued… R e s e a r c h i n B r i e f

Issues and Findings The evaluation found significant favorable resources available in each site. (See continued… effects of Weed and Seed on key outcome “Study Methodology.”) In general, the sites measures for some sites and time periods. were organized along these lines: Key findings: The effectiveness of The evidence is modest in terms of its weeding and seeding activities Grantee organization. The grantees varied across the eight sites. The statistical significance, but the indicators evaluation found: consistently point in favorable directions. included mayors’ offices, local police At the same time, the evaluation has departments, and local nonprofit ● Preexisting community features 501(c)(3) organizations. The staff as- may make Weed and Seed easier pointed out a number of weak links in the or more difficult to operate effec- chain, most noticeably the limited and signed to the Weed and Seed effort and tively. Important factors included tenuous role that many local prosecutors their ability/authority to operate cohe- the strength of the social and played in the weeding process. A number sively in conjunction with community institutional infrastructure (an representatives were among the most established network of community- of local prosecutors reported they simply based organizations and commu- lack the funding and personnel to conduct important factors in successful program nity leaders), the severity of crime enhanced prosecution of the target area implementation at the eight evaluation problems, geographical advantages caseloads generated by more aggressive sites. In general, an adequate supply of favoring economic development, dedicated staff resources, from both and transiency of the community policing activities. Although difficult to population. measure with precision, the effect of this grantees and other participating agencies, constraint almost certainly has been to limit produced positive results in program ● The mix of weeding and seeding the removal of offenders from target areas. implementation, oversight, and cohesion. activities and the sequencing of In most sites, the role actually played by these components appear to be the grantee was dynamic, as key partici- important factors in gaining com- Setting up Weed and Seed at munity support for the program. pants changed, programs and roles Important positive factors included the local level evolved, and community and interagency early seeding, sustained weeding, high-level task forces combined The eight sites differed substantially in relationships developed. with community policing, and an how they organized their Weed and Seed active prosecutorial role. programs, due, in large part, to the unique The U.S. Attorney’s role. The U.S. law enforcement and social needs and Attorney is asked to play a central role in ● Sites appeared to have greater organizing the steering committee and success if they concentrated their existing organizational infrastructure and program resources on smaller population groups, especially if they could similarly channel other Exhibit 1. Number of Weed and Seed sites and annual funding public funds and also leverage private funds. Fiscal Year Number of Funded Sites Total Program Funding (millions) ● A less tangible ingredient that seemed to characterize the more 1991 3 $0.5 successful programs was the active and constructive leadership of key 1992 20 $11.5 individuals. 1993 21 $13.5 ● The most effective implementa- 1994* 36 $31.5 tion strategies were those that relied on bottom-up, participatory 1995* 36 $32.5 decisionmaking approaches, espe- 1996* 88 $37.5 cially when combined with efforts to build capacity and partnership 1997* 118 $37.5 among local organizations. 1998* 176 $42.5 The full cross site analysis and 1999* 200 $49.0 the eight case studies are also available. * Since 1994, in addition to appropriated funds, the U.S. Department of Justice has allocated $9 million annu- Target audience: Congressional ally in Asset Forfeiture Funds for Weed and Seed-related task forces administered through the Drug Enforce- representatives and legislative staff; ment Administration, the Federal Bureau of Investigation, and other DOJ law enforcement agencies. Federal, State, and local law en- forcement officials; State and local government officials; and criminal justice practitioners and 2 researchers. R e s e a r c h i n B r i e f

What Is Weed and Seed? W eed and Seed is a strategy to The Federal oversight responsibility for Notes mobilize and coordinate resources in the each participating site rests with the U.S. a. Executive Office for Weed and Seed, Opera- targeted communities rather than simply Attorney’s Office for the corresponding dis- tion Weed and Seed, Implementation Manual, a program or mechanism to fund local trict. This decentralized arrangement was in- Washington, D.C.: U.S. Department of Justice, activities that share no collective aim. tended to reinforce local participation while Office of Justice Programs, Executive Office for Weed and Seed is an “incubator for social at the same time providing a more hands-on Weed and Seed: 1–3 and 1–4. change” to stabilize the conditions in Federal role—in particular, to enable Federal high crime communities and to promote prosecutorial action to be more responsive b. The Executive Office for Weed and Seed community restoration.a The key compo- to local law enforcement initiatives. Other regards community policing as “increasing nents of this strategy are as follows: components of each local organization’s police visibility and developing cooperative structure generally include: relationships between the police and citizenry ● Enhanced coordination—Coordi- in target areas.” The associated techniques nated analysis and planning of local prob- ● The Weed and Seed steering committee, include foot patrols, police ministations, nui- lems and strategies to address them. which establishes operational goals, designs sance abatement, victim referrals to support services, and community relations activities, in and develops programs, guides implementa- which the community is encouraged to under- ● Weeding—Concentrated and tion, and assesses program achievements. enhanced law enforcement efforts to take such initiatives as neighborhood watches, citizen marches and rallies, drug-free zones, identify, arrest, and prosecute violent ● The weeding committee, which plans and and graffiti removal. See Executive Office for offenders, drug traffickers, and other monitors the law enforcement efforts, Weed and Seed, Operation Weed and Seed, criminals operating in the target areas. including interdiction and prosecution. Implementation Manual: 9–6. The objective is to remove criminals from the target areas. ● The seeding committee, which plans and c. Executive Office for Weed and Seed, Weed monitors the prevention, intervention, treat- and Seed Fiscal Year 1998 Program Guide and ● Community policing—Proactive ment, and neighborhood restoration efforts. Application Kit for New Sites, Attachment 2, police/community engagement and prob- Washington, D.C.: U.S. Department of Justice, lem solving in which police officers are as- ● The Weed and Seed program staff, Office of Justice Programs, Executive Office for signed to specified geographic locations. who operate and maintain daily program Weed and Seed, 1998: 2–1. In its literature, By gaining the growing trust and support activities. the Executive Office for Weed and Seed de- of the community, police and prosecutors scribes safe havens as an integral part of a Each site is required to create at least one “risk factor and protective factor” approach to engage residents and businesses as prob- “safe haven,” a highly visible and accessible crime prevention. This is viewed as the coun- lem-solving partners in the law enforce- multiservice center where youths and adults terpart to comprehensive community-based ment effort. This effort is the bridge can receive needed services, develop rela- disease prevention programs, which have op- between weeding and seeding.b tionships, enhance personal skills, and erated effectively in the public health arena. In the Weed and Seed context, risk factors, those ● Seeding—Human services—including find opportunities to be productive and that make an individual susceptible to criminal successful.c Special effort is made to keep afterschool, weekend, and summer youth behavior, must be identified and addressed. these safe havens secure from drug traffick- activities; adult literacy classes; and paren- At the same time, protective factors, those tal counseling—and neighborhood revi- ing and other criminal activities. that serve to counter or neutralize risks, must talization efforts to prevent and deter be enhanced. The full National Evaluation Research Re- further crime. port is available in print and online. The eight case studies are available online.

3 R e s e a r c h i n B r i e f bringing together community organiza- nals, especially those engaging in drug goals and objectives, provided guid- tions and other Weed and Seed partici- trafficking and violent crime. By provid- ance and oversight on key program de- pants. A basic Weed and Seed premise ing more effective crime detection and sign and implementation issues, and is that the U.S. Attorney’s Office pro- response, speedier investigations and integrated weeding and seeding at the vides Federal oversight and coordinates trials, and the stricter sentences avail- policy level. In most sites, the steering Federal, State, and local law enforce- able through Federal prosecution, po- committee played a critical role in ment and prosecutorial activities, as lice and prosecutors seek to get drug coordinating efforts across agencies, well as general DOJ oversight of the dealers and other criminals off the sectors, and jurisdictions. For several Weed and Seed strategy. Through such street. The goal of these activities is to sites, the steering committee or Weed coordination, some sites effectively build trust and support with the resi- and Seed community organizations used Federal law in weeding strategies dential and business communities. provided a critical means of resident and mobilized resources for seeding Police and prosecutors can then engage participation in program decision- programs from a variety of Federal area residents and business people as making. Across all eight sites, steering agencies. In other sites, however, Weed problem-solving partners, who, through committee members included key pub- and Seed was managed and operated by their cooperation, will promote further lic agency representatives and local a city agency (e.g., a police depart- arrests and prosecutions. government officials and various social ment), and the role played by the U.S. service providers, community-based Attorney was much more limited. Steering committee. The role and organizations, and residents. composition of steering committees var- Police and prosecutors. Additional ied among sites, with some dominated Interorganizational linkages. law enforcement resources are brought by public-sector representatives and Increased interagency collaboration to bear on targeted areas. Police and others more heavily represented and was a central component of most weed- prosecutors concentrate their efforts to guided by community residents. ing strategies. Such efforts included identify, arrest, and prosecute crimi- In general, the committees established monthly meetings between Weed and

Exhibit 2. Target area characteristics

Site/Target Area(s) W/S Start Funding Part I Area in Date Crime Rate* Square Miles Population

Akron: West Side October 1995 $158,172 69.5 3.5 23,930

Hartford: Stowe Village January 1995 $2,448,424 199.2 0.001 1,300

Las Vegas: $2,168,066 Meadows Village October 1994 206.3 0.05 4,819 West Las Vegas October 1994 118.0 1.6 10,103

Manatee/Sarasota: $2,515,000 North Manatee October 1994 73.9 1.1 3,327 South Manatee October 1994 85.3 1.2 8,620

Pittsburgh: Hill District April 1992 $3,275,000 246.6 0.5 4,244

Salt Lake City: West Side March 1995 $550,000 55.1 6.3 22,000

Seattle: Central District January 1993 $3,125,000 172.6 1.2 13,812

Shreveport: Highland/Stoner Hill February 1995 $2,250,000 211.6 4.6 12,668

*Per 1,000 residents in the year preceding implementation of Weed and Seed.

4 R e s e a r c h i n B r i e f

Study Methodology T he National Evaluation of opera- fewer questions than the 1995 survey. The participants in each site (excluding Akron, tion Weed and Seed was designed to decision to proceed in 1997 with telephone Ohio) to gain their perspectives. Evaluation gain the perspective of target area resi- interviewing and a shorter questionnaire was staff visited 4 to 6 programs in each of dents in eight sites, to assess the trends in based on the difficulties experienced in 1995 the sites and conducted interviews with arrests and crime in those areas, and to in completing the targeted number of inter- groups of 7 to 10 people. These interviews draw from the personal experiences of views per site. were not intended to measure outcomes participants in the seeding programs. The of the programs that were visited; rather, evaluation included a wide variety of ac- To analyze arrest and crime patterns in the they were designed to learn the perceived tivities at each site as follows: 10 Weed and Seed target areas, the re- benefits and drawbacks of the programs search team requested computerized inci- from the individuals who participated in ● A review of funding applications and dent-level data from law enforcement them. Long-term effects of these pro- other significant program documents. agencies, including basic facts about each grams on the lives of the program partici- arrest, crime (e.g., when and where the pants cannot be deduced from the ● Individual interviews with key program crime was committed), and associated crimi- interviews conducted. administrators, senior law enforcement nal charges. Because the primary interest staff, managers of seeding organizations was in arrests that were made and crimes The seeding programs that were visited and activities, service providers (both cur- that occurred in the target areas, procedures fall under four general categories: youth rent and former), and community leaders. were developed to identify those arrests and recreation and education, community crimes from data files provided by each juris- building, employment and training, and ● Analysis of automated, incident-level diction. In most cases, the procedures in- violence prevention. Programs visited crime and arrest records provided by the volved geocoding the address where the included Police Activities (or Athletic) local police departments. arrest was made or where the crime was Leagues, Boys and Girls Club Athletic Pro- ● Group interviews with seeding pro- committed. grams, Safe Haven After-School Programs, gram participants. and Youth Enrichment Programs. Based The focus has also been on the seven Part I on the comments of the participants inter- ● Resident surveys in target areas, con- crimes—homicide, rape, robbery, aggra- viewed, the seeding programs appeared to ducted in June 1995 by the Institute for vated assault, burglary, larceny, and auto provide services that would otherwise not Social Analysis and in December 1997 by theft. Although these crimes account for have been available in the target areas. Abt Associates Inc. only one segment of the overall crime prob- Most of the interviewees also indicated lem (ignoring, for example, order mainte- participation in the seeding programs has Resident surveys. To a large extent, nance and other crimes closely associated been a positive experience that helped both surveys focused on the same is- with quality-of-life issues), there are standard them feel more secure emotionally, physi- sues—resident perceptions of crime, definitions of these crimes that all law en- cally, or both. The general themes that public safety, police performance and forcement agencies follow, thus allowing for emerged focused on providing additional activities, quality of life, and awareness site-to-site comparisons. Given that control- structure and discipline in the lives of of the Weed and Seed program. There ling drug trafficking and drug-related crime target area youths and providing opportu- were, however, some differences in the is one of the key Weed and Seed objectives, nities and assistance for adults to work methods used in the two surveys. The the researchers also focused on drug arrests. toward personal and professional growth. 1995 survey consisted of 1,531 in-person interviews, while the 1,995 interviews in Onsite observation of programs. Inter- 1997 were conducted by telephone. In views were conducted with seeding program addition, the 1997 survey consisted of

5 R e s e a r c h i n B r i e f

Seed officers and probation/parole they felt more connected to the com- employed a bottom-up, grassroots officers to catch parole violators, fre- munity and service providers. A direc- approach tended to build trust among quent joint Federal and local special tor of a youth service organization in residents and community-based lead- narcotics operations, and the creation Shreveport said: ers and enhanced community capacity of formal interagency Weed and Seed for crime prevention and reduction law enforcement task forces, such as Establishing partnerships is the key and social development. in Las Vegas, Manatee/Sarasota Coun- to Weed and Seed—cooperation on ties, and Pittsburgh. In Salt Lake City, projects and working together to Weed and Seed also provided a forum Weed and Seed worked closely with bring the community together. for leadership development among the existing Metro Narcotics Task [You] can touch more people as a residents. In Salt Lake City, for ex- Force, which focused on high-level group of organizations than as a ample, a resident took the lead in the drug dealers. single entity. initial Weed and Seed grant applica- tion and played an instrumental role The eight Weed and Seed sites tried to Building trust and in early program implementation. In build their seeding programs around community capacity the North and South Manatee target existing resources, in addition to creat- areas, charismatic individual leaders ing new partnerships. For example, Several of the evaluation sites encoun- emerged who essentially ran the safe Hartford Weed and Seed integrated tered early community resistance to havens as volunteers and directed a seeding efforts with the existing U.S. Weed and Seed because residents variety of community activities. For all Department of Housing and Urban were concerned about an exclusive sites, it remained a central organiza- Development (HUD) Family Invest- focus on enforcement or the potential tional challenge to continuously ment Center, and the Salt Lake City for targeted harassment. The clear develop and broaden leadership to Weed and Seed program built on the lesson from these experiences was the implement and sustain the Weed and Comprehensive Communities Program. importance of involving residents early Seed strategy. Extensive partnerships and collabora- in Weed and Seed planning, providing residents with substantial program tions were created through Weed and Approach to law enforcement Seed in the public and private sectors, authority, and nurturing higher levels with most of the private-sector partner- of interaction and trust between pub- Most sites developed and implemented ships with nonprofit and community- lic-sector representatives and those coherent law enforcement strategies based organizations. In Shreveport, for they serve. The seeding component that responded to local conditions and example, Weed and Seed arranged for and community policing were typically incorporated stronger patrols at the existing services to come to the target the intended means for building com- street level with some degree of higher area, including free immunizations for munity trust and encouraging partici- level interagency cooperation. Law en- children provided by the State Depart- pation. In the initial concept, it was forcement approaches across the target ment of Public Health, a library book- assumed these elements would follow areas typically included increased mobile, and free computer training intensified enforcement. In practice, police presence through additional from a local university. In Seattle, the it became apparent to many Weed staffing and overtime, with the major- Washington Insurance Council joined and Seed organizers that the target ity of sites assigning officers dedicated Weed and Seed to establish the Seattle areas had to become involved at the to the target area, and increased spe- Neighborhood Action Program as a outset and a comprehensive strategic cial operations for targeted law en- public/private partnership to help revi- plan was needed to bring this about. forcement, especially for drug-related and violent crime. talize a Weed and Seed neighborhood. Even when target areas had strong Seeding program partners and provid- preexisting community organizational Multiagency task forces. The sites ers who were interviewed emphasized infrastructures, considerable resources developed varying degrees of increased how Weed and Seed increased coordi- were needed to effectively catalyze local, State, and Federal coordination, nation and communication links resident participation and increase the whether in targeting offenders, narcot- among neighborhood groups and other organizations’ scope of operations and ics operations, prosecution, or proba- agencies. Almost all respondents said outreach. Weed and Seed sites that tion/parole. Local responses ranged

6 R e s e a r c h i n B r i e f from increasing communication through residents knew the four dedicated In Las Vegas, for example, the U.S. monthly meetings to creating formal in- “bumble bees”—the bicycle patrol Attorney’s Office played both roles teragency and multijurisdictional task officers in their yellow shirts and black during the early stages of program force operations housed at the same fa- shorts—and, in Pittsburgh, residents implementation. The role played by cility. Although multiagency task forces lobbied to keep one of their Weed and the U.S. Attorney in each area might concentrated on the target area, they Seed officers from being reassigned to be strong or weak, depending on the pursued drug cases across jurisdic- another district. One police lieutenant personalities and motivations of the tional lines. Consequently, the benefits in Las Vegas would find jobs for gang partners involved. of these efforts extended beyond the members, and, occasionally, he would target areas, particularly when the fo- drive them to their jobs or check with Prosecution. Overall, prosecution cus was on high-level drug dealers who their supervisors to see how the youths has been a relatively weak link in controlled large operations. Task force were doing on the job. Although weed- Weed and Seed due to various institu- missions varied primarily according to ing has typically involved less resident tional, political, and judicial issues. the nature of crime in the target areas participation than has seeding, com- In the majority of sites, there was no and preexisting law enforcement opera- munication between residents and the special Federal or local prosecution tions, so the Weed and Seed task forces police seems to have increased across or tracking of Weed and Seed cases complemented existing efforts. sites. except for efforts conducted by joint Federal/State/local task forces. In gen- The community policing bridge. In addition to enhancing community eral, district attorneys operate with The implementation of Weed and policing efforts, Weed and Seed limited resources and in politicized Seed enabled most sites to expand or provided a vehicle for mobilizing resi- environments that act as barriers to the strengthen community policing efforts dents to participate in crime preven- provision of the additional resources or institute new programs; better con- tion and, in some cases, for creating needed for local prosecution of Weed centrate, coordinate, and integrate effective structures for community and Seed cases. Some local prosecu- efforts within police departments; and authority and leadership. Responses tors said although weeding activities increase integration of law enforce- ranged from increasing neighborhood generated more criminal complaints, ment with seeding type activities. watches, to community meetings, to a their office budget was not increased Community policing initiatives such citizens’ advisory committee that pro- to hire new prosecutors or to cover the as nuisance abatement, landlord pro- vided guidance on law enforcement operating costs that would be added grams, graffiti eradication, code en- priorities. by enhanced prosecution of Part II or forcement, and neighborhood cleanups quality-of-life offenses. helped improve property maintenance The U.S. Attorney’s role. At the and neighborhood environments. Federal level, the U.S. Attorney’s two At the local level, police departments Officers participated in a wide range potential roles were to serve as: and prosecutors often work through of youth recreation and education different political systems. Most ● A coordinator or “people mover” programs in the target areas. Such police departments are city based and for Weed and Seed operations. For receive their local and Federal funding activities engaged local youths in example, when the U.S. Attorney constructive activity, provided positive through their city council. That pro- chaired steering committee meet- cess is fairly well defined in most cit- role models for the youths, and built ings and used resources of the community relations. ies. Prosecutors, on the other hand, Office of the Attorney General in often are county based and receive The assignment of dedicated officers the process, this served as a stimu- their funding from the county. This to the Weed and Seed target areas was lus to other partners. funding process often encompasses important in building relationships ● A key player in multijurisdictional multiple municipal priorities and with residents and in aiding enforce- task forces when the focus was on other competing political consider- ment through better knowledge of high-level drug distribution and ations. The role of Federal prosecution the neighborhood, better intelligence, sales. This helped to integrate, also varied from no or few Federal and the ability to operate proactively. coordinate, and focus local and prosecutions to substantially increased In Shreveport, for example, many Federal enforcement efforts. Federal, State, and local cooperation.

7 R e s e a r c h i n B r i e f

Approach to seeding with varying degrees of program 24 percent; North Manatee, 18 per- breadth, depth, and duration. In sev- cent; the Shreveport target area, 11 One of the greatest challenges for the eral sites, providing communities with percent; the Central District in Seattle, sites was to develop an appropriate a voice in the seeding grant award pro- 10 percent; and West Las Vegas, seeding strategy with community cess was critical to gaining community 6 percent. Three target areas experi- members that targeted Weed and Seed participation and trust. enced increases in Part I crimes: resources most effectively and lever- South Manatee, 2 percent; Meadows aged existing public and private Seeding initiatives undertaken by the Village in Las Vegas, 9 percent; and resources. Simply selecting and suc- evaluation sites can be clustered as Salt Lake City, 14 percent. A compa- cessfully implementing seeding pro- follows, in order of predominance: rable estimate was not possible for the grams, such as safe havens, in itself Akron target area due to insufficient was more difficult for most sites than ● Prevention and intervention pro- data. implementing their weeding programs, grams for youths that included with some sites initially stumbling at afterschool programs, safe During this same time period, in seven this level of program execution. havens, recreation and sports target areas—Hartford, Pittsburgh programs, skills and employment (Crawford-Roberts), North Manatee, Seeding was inherently a broader and training, job development, and South Manatee, Shreveport, West Las more complex task, both in develop- health and substance abuse-related Vegas, and Salt Lake City—Part I ment of goals and strategies and from programs. crime rates declined more or increased a practical organizational standpoint. ● less than in the rest of the city or Seeding efforts involved engaging par- Neighborhood restoration, such as county. ticipation and commitment from pub- neighborhood cleanups and code enforcement. lic- and private-sector organizations, A relationship appears to exist be- whereas weeding had a relatively ● Community building and commu- tween crime trends and the concentra- clearer mission, operating within more nity development initiatives. tion of program resources in sites that established hierarchical structures of had the largest increases or decreases law enforcement and criminal justice ● Adult employment and economic advancement programs, such as in crime. Hartford, for example, has organizations. Due to the broader and the smallest target area in terms of less defined nature of seeding, rela- computer, Internet, and educational training. population and area, while Salt Lake tively more time was needed for City has the largest single target area planning, relationship building, and ● Family support services targeted to in square miles and, along with Akron, gaining consensus and commitment adults. the smallest level of Federal Weed and from the wide range of participants ● Community economic development. Seed funding. (See exhibit 2.) who shaped this domain. Seeding, by its nature, is resistant to short-term Finally, changes in the drug arrest Impact on crime trends fixes designed to produce observable, rates appear to follow the same general overnight results. Across the evaluation sites, crime pattern as the changes in the Part I patterns varied widely. As exhibits crime rate. For example, among those The eight Weed and Seed sites tried to 3a and 3b show, in nine target areas six target areas for which there are build their programs around existing available data allowed a comparison of arrest data, the four with decreases resources, in addition to creating new the number of Part I crimes (homicide, in Part I crime from the year prior to partnerships. Seeding program part- rape, robbery, aggravated assault, bur- Weed and Seed through the second ners and providers who were inter- glary, larceny, and auto theft) in the year of implementation (i.e., Hartford, viewed emphasized how Weed and year prior to program implementation Pittsburgh, North Manatee, and Seed increased coordination and com- to the second year of Weed and Seed. Shreveport) all experienced initial munication links across neighborhood Six of these areas showed declines: high rates of drug arrests—suggesting groups and other agencies. Weed and Stowe Village in Hartford, 46 percent; an initial period of intense weeding Seed sites reflected different emphases Crawford-Roberts (one neighborhood activities—followed by declining in funding local seeding programs— within the Hill District) in Pittsburgh, drug arrest rates. Assuming the level

8 R e s e a r c h i n B r i e f of enforcement as measured by police Exhibit 3a. Percent change in Part I crime from the year preceding Weed & presence has remained somewhat con- Seed to year 1 and to year 2: target areas stant, this trend reflects success in re- ducing drug activity. However, the Salt 50 Lake City target area and South Mana- tee both experienced large increases 40 in the number of drug arrests in 1997 compared with 1996, suggesting per- 30 West Side (Salt Lake City)* haps these sites had not yet succeeded in reducing the level of drug activity 20 in the target areas. In the case of Salt South Highland/ 10 Stowe Village West Manatee Stoner Hill Lake City, an influx of gang activity is (Shreveport) an important contextual factor, raising (Hartford) Las Vegas the question of whether the crime rate 0 would have been even higher there -10 Meadows without Weed and Seed. Village West Side Central District (Las Vegas) (Seattle) -20 (Akron) North Assessing the Weed and Manatee Seed strategy -30 Crawford-Roberts (Pittsburgh) The evaluation investigated two major -40 questions: What factors appeared to year 1 year 2 help or inhibit successful implementa- tion of Weed and Seed; and did detect- -50 able changes in outcome measures * Year-to-year comparisons in Salt Lake City are based on 5-month periods only. Weed and Seed take place as a consequence of Weed started in August 1995, and data are available only back to March 1995. and Seed?1 Exhibit 3b. Percent change in Part I crime from the year preceding Weed & Effects on crime and public Seed to year 1 and to year 2: comparison areas perceptions Regarding the specific survey-reported 15 pattern of residents’ perceptions of Salt Lake City* 10 crime, public safety, and police perfor- Manatee County mance, the target areas cluster into 5 three groups: Hartford Shreveport 0 ● North Manatee and Pittsburgh exhibited substantial evidence of Las Vegas -5 changes in residents’ perceptions Akron across multiple outcome measures, -10 including the severity of crime and Pittsburgh police effectiveness in controlling Seattle -15 crime.

● Akron, Hartford, and Seattle exhib- -20 year 1 year 2 ited some evidence of changes in residents’ perceptions on selected -25 crime dimensions, either drug- * Year-to-year comparisons in Salt Lake City are based on 5-month periods only. Weed and Seed related crime (Akron and Seattle) started in August 1995, and data are available only back to March 1995.

9 R e s e a r c h i n B r i e f

or violent and gang-related crime What factors appear to have promoted funding to a narrowly targeted group of (Hartford). successful implementation of the residents in a small geographical area) program and thus have promoted the and channeling and leveraging other ● Las Vegas, Salt Lake City, and achievement of the program’s intended funds. Shreveport exhibited little evidence results? To address this question, the of changes in residents’ perceptions evaluation considered site characteris- Leadership and partnership. A less of general public safety or the tics and program features. tangible ingredient that characterized severity of specific types of crime the more successful programs was the in the neighborhood. Community setting. Preexisting active and constructive leadership of Based on the pattern of findings on community features may make Weed key individuals. By its very nature, the rate of Part I crimes as well as and Seed easier or more difficult to Weed and Seed places a great pre- responses to crime-related survey operate effectively. Important factors mium on effective coordination of questions, the target areas fall into included the strength of the social and groups with different organizational four categories, first according to the institutional infrastructure (an estab- missions, responding to different evidence of reduced Part I crimes and lished network of community-based constituencies. To establish effective then (within each category) according organizations and community leaders), working relationships among these or- to the evidence of improved public the severity of crime problems, geo- ganizations required personal energy perceptions: graphical advantages favoring eco- and initiative and an organizational nomic development, and transiency of structure that facilitated interaction. ● Pittsburgh and Hartford showed the community population. The last For example, those sites that based strong evidence of reduced Part I factor is particularly important. Rapid weeding and seeding staff in the same crimes and improved public per- turnover, as in the case of Meadows facility tended to develop better coop- ceptions on crime-related measures Village in Las Vegas, undermines the eration and coordination between the (e.g., reduction of fear of crime, crime-fighting effect of community two program components. public safety). policing and makes it difficult, if not impossible, to generate the ongoing The most effective implementation ● Manatee/Sarasota (North Manatee) community involvement that is a key strategies were those that relied on and Shreveport showed substantial bottom-up, participatory decision- evidence of reduced Part I crimes. element in the Weed and Seed concept. making approaches, especially when (North Manatee also showed im- combined with efforts to build capacity proved public perceptions on mul- Program design. The mix of weeding and partnership among local organiza- tiple crime-related measures.) and seeding activities and the sequen- tions. This required a long-term per- ● Seattle, Akron, Las Vegas (West cing of these components appear to be spective about the program and its Las Vegas), and Manatee/Sarasota important factors in gaining community potential to bring about community (South Manatee) showed some evi- support for the program. Important change. Such sites, including some dence of reduced Part I crimes. positive factors included early seeding, that achieved substantial crime reduc- (Akron and Seattle also showed sustained weeding, high-level task tions within the time period analyzed, improved public perceptions.) forces combined with community po- have established a stronger foundation licing, and an active prosecutorial role. and more sustainable basis for further ● Salt Lake City and Las Vegas community-targeted initiatives. (Meadows Village) showed no evi- Concentration of funds. As is to dence of reduced Part I crimes. be expected, sites appeared to have Policy implications greater impact on crime rates if they Factors favoring successful concentrated their program resources In charting the future direction of implementation of Weed on smaller population groups, espe- Weed and Seed, policymakers have a and Seed cially if they could similarly channel number of strategic choices to make. other public funds and also leverage These include designating sites for Of course, these findings must be in- private funds. The important factors continued funding, selecting sites terpreted in terms of local circum- included funding “intensity” (applying for new awards, and allocating funds stances and program approaches.

10 R e s e a r c h i n B r i e f among participating sites over time. (See “Where is Weed and Seed go- ing?”) The key policy question is how Where Is Weed and Seed Going? to use program funds most effectively in ways that make the greatest long- B eginning in 1996, the Weed and Funds, Asset Forfeiture Funds, COPS hiring term contributions to controlling crime Seed program altered its paradigm and awards) as well as other Federal, State, and promoting a safe living environ- approach to site longevity and funding local, and private resources. The long- ment for residents. Other nonfunding levels. After funding 36 sites as part of a range objective is for recipient sites to use considerations include improved coor- 3-year demonstration at approximately the Weed and Seed funds to develop suc- $750,000 per site, funding was extended cessful, self-sustaining community-based dination among Federal, State, and to additional sites at a lower level. By interventions. local public and private partners. 1997, Weed and Seed funding was offered to almost 120 sites, typically at Consistent with the program’s emphasis The experience of the eight sites $250,000 per year. At this new, lower on flexibility and customer orientation, evaluated suggests Weed and Seed has annual funding level per site, Weed and sites are given considerable latitude to de- affected the target areas through either Seed appropriated funds to 176 sites in velop programs with the Executive Office (or both) of two avenues. The first, 1998. In 1999 the program grew to ap- for Weed and Seed grant funds and to program effectiveness, relates to the proximately 200 sites, despite a drop in request training and technical assistance. specific initiatives that focus on law total national funding. Increasingly, experienced sites are asked to enforcement and crime prevention. provide peer-to-peer training to show new These activities, both on the weeding In fiscal year 2000, new policies will re- and developing sites what to do and what quire the original 36 sites to submit com- to avoid to develop successful local pro- and seeding sides, appeared to have prehensive reapplications to be eligible grams. Sites are encouraged to organize varying degrees of success in reducing for funding. Subsequently, every site regional training sessions to facilitate these crime and improving perceptions of must reapply after 5 years of continuous exchanges and coordinate with other public safety. The second, community funding. This new approach is consistent Federal agency partners. mobilization, is the process in which with the basic Weed and Seed strategy of Weed and Seed provides a catalyst for pursuing effectiveness through coordina- In addition to providing funding directly greater involvement of neighborhood tion and leveraging of additional public to sites, Weed and Seed supports many multisite activities. For example, in coop- residents and community-based orga- and private resources (in many cases, sev- eration with the U.S. Navy Drug Demand nizations. As mentioned earlier, some eral times the amount of Weed and Seed funding). Reduction Task Force and other military sites were more predisposed than oth- partners, Weed and Seed supports the ers to participatory problem-solving Additional resources sites are urged to implementation of the Drug Education for arrangements by virtue of their preex- investigate include other U.S. Department Youth (DEFY) program in Weed and Seed isting infrastructure and active leader- of Justice funding streams (e.g., Local communities. Weed and Seed DEFY sites ship by respected individuals and Law Enforcement Block Grants, Byrne increased from 3 in 1996 to 60 in 1999. established organizations. Formula Grants, Juvenile Justice Formula

These two mechanisms of change are clearly interrelated. The experience of Seed influence. To maximize impact tegic effort is also an important the sites studied here showed: on crime rates, Weed and Seed means of increasing the intensity of should seek the highest feasible con- the intervention. Increasing coordina- ● The Weed and Seed program has centration of funds in the program tion with other Federal funding been a strong stimulant to community sites. Given the annual funding con- sources is another key element. Fur- coalition building. Public and private straint of the congressional appro- thermore, the evaluation finds that organizations came together, for the priations process, this implies a more Weed and Seed funding has acted first time in a number of sites, to de- selective process in choosing sites to as a significant catalyst for general velop interventions that would have a receive new awards and/or limiting community revitalization efforts and broad base of support. It seems clear the number of years that ongoing that most target area communities these developments would not have sites receive program funding. Hav- have undertaken programs and occurred in the absence of Weed and ing a concentrated and focused stra- created beneficial community

11 R e s e a r c h i n B r i e f

organizations that likely would not successful ones—on these issues Social Analysis identified such comparison have come into existence without are, to a large degree, widely appli- sites and included them in the 1995 survey of community residents and other baseline data Weed and Seed. cable. New sites should receive collection efforts. Soon thereafter, however, NIJ the benefit of this experience. This and Abt Associates concluded the comparison ● In selecting sites for new program seems especially important in pro- sites were not sufficiently comparable to their funding, Weed and Seed should moting the partnership arrange- corresponding target areas in demographic place its funding priority on sites ments that have characterized the characteristics and crime trends. The matched with geographically small target site approach was not pursued further. See more successful programs examined areas and with favorable commu- Terence Dunworth, et al., Overview of Institute in this research. for Social Analysis National Evaluation nity settings and program designs. Baseline Data and Implications of the Data for This is more likely to produce de- Note the Weed and Seed Impact Evaluation, Cam- monstrable successes. In turn, this bridge, MA: Abt Associates Inc., January 1996. should increase potential spinoffs 1. The most rigorous means of establishing from the Weed and Seed target area the counterfactual is an experimental design in Findings and conclusions of the research which the intervention, in this case Operation reported here are those of the author(s) and do to other areas in the site. Weed and Seed, is not implemented among a not necessarily reflect the official position or randomly selected set of subjects, in this case policies of the U.S. Department of Justice. ● The Executive Office for Weed and the sites. Such a design was infeasible in this Seed should expand its provision of context. Another approach would have been to The National Institute of Justice is a technical assistance to the funded match each evaluation site with a comparison component of the Office of Justice sites. The lessons learned from the site of similar baseline characteristics. At an Programs, which also includes the Bureau more successful sites—and the less earlier stage of this evaluation, the National of Justice Assistance, the Bureau of Justice Institute of Justice (NIJ) and the Institute for Statistics, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime. Terence Dunworth, Ph.D., is man- The research for this study was sup- aging vice president and project ported under grant number 95–DD– This document, the full cross site analy- director at Abt Associates Inc. BX–0134 by the National Institute of sis, the eight case studies, and other NIJ Gregory Mills, Ph.D., is a senior Justice, Office of Justice Programs, publications can be found at and down- associate at Abt Associates Inc. U.S. Department of Justice. loaded from the NIJ Web site (http:// www.ojp.usdoj.gov/nij). NCJ 175685

U.S. Department of Justice PRESORTED STANDARD Office of Justice Programs POSTAGE & FEES PAID DOJ/NIJ National Institute of Justice PERMIT NO. G–91

Washington, DC 20531

Official Business Penalty for Private Use $300