The National Faith-Based Initiative (NFBI) began in late 1998. In the 12 months from September 2001 through August 2002, 12 sites began participating in the NFBI. The programs differed in size, theological orientation, and strategy; however, they all focused on high-risk youth; included components on mentoring, education, or employment readiness; and involved collaborations among faith-based institutions, justice institutions, and social service agencies. This report covers the first phase of the multiyear evaluation, which focused on the implementation of programs and site strategies, as well as the successes and challenges in setting up programs. It does not encompass a final assessment of the value and effectiveness of faith-based mentoring programs; such research is currently being undertaken. By fall 2001, four sites had developed a mentoring program and had organized the recruitment, screening, training, and monitoring components of the program. Although each of these programs emphasized mentoring, the design of the programs and the array of additional services differed. The programs are located in Baton Rouge, LA; Bronx, NY; Brooklyn, NY; and Philadelphia, PA. In addition to a survey of all volunteers active in these programs during a 3-month period (March-May 2002), researchers visited the Philadelphia and Brooklyn programs three times and the Baton Rouge program once in order to interview staff, mentors, and youth. Group mentoring activities were observed. Information was obtained from program officers who visited the sites bimonthly and talked with staff monthly. Program officers had conducted a yearly intensive assessment of the sites' organizational capacity and program implementation. In addition, the sites submitted quarterly progress reports to Public/Private Ventures (P/PV) that documented implementation efforts as well as demographic information and the status of each mentoring match. The evaluation derived three main lessons that can guide other faith-based organization or funders who are implementing similar programs. First, drawing on faith-based volunteers requires a different approach to the screening, training, and monitoring practices common to mentoring programs. Staff had to adapt their screening, training, and monitoring to address the possibility of proselytizing by the mentors. The sites discussed this issue with volunteers to ensure they were comfortable mentoring without engaging in inappropriate faith-sharing. The sites also discussed appropriate ways to share their faith with the youth. They emphasized that mentors could not pressure or require youth to engage in any faith-related activities; mentors were allowed to invite youth to church and church-related events. The youth felt the mentors were respectful of their limits and did not "push" faith on them. Second, the size and composition of the local religious community set boundaries and provided opportunities for faith-based organizations' recruitment efforts. This allowed NFBI staff to recruit mentors by relying on the support of pastors from local congregations, who often made direct appeals for mentors from the congregation. Third, attention must be given to maintaining the relationship when mentoring older, high-risk youth, although this may require additional resources, especially for case management. 8 tables, 19 notes, and appended survey questionnaire and data on the duration of the mentoring relationships by site