Leading school change is a challenging endeavor. Successful leadership requires strategic and sustained effort, particularly in the shifting and uncertain environment of urban public schools. The concept of distributed leadership--in which multiple actors tackle the challenges of school leadership in concert--is a promising way to strengthen professional practice and thereby improve the educational experiences of all students. The Annenberg Distributed Leadership (DL) project was one of the first efforts in the nation to deliberately take on the challenge of designing and implementing a concerted effort to build distributed leadership capacity in a diverse set of urban schools to improve the quality of teaching and learning. The DL project featured the careful selection of leadership team members to identify and lead instructional improvement efforts; ample professional development to build cohesive teams and help members understand the motivational, psychological, and pedagogical aspects of advocating instructional change; resources to apply to the task; and ongoing school-based coaching to guide leadership team efforts. This report describes the Consortium for Policy Research in Education's mixed-method evaluation of the DL project. The evaluation featured a cluster randomized control trial, where schools first agreed to participate in the study, and then were chosen by lottery to participate in the DL project or serve in the comparison group. Overall there were 16 DL schools and 21 comparison sites in the evaluation. Central evaluation findings included: (1) The DL project successfully identified and developed leadership teams that were significantly higher functioning than were leadership teams in the comparison schools; (2) Principals and team members forged new working relationships that productively expanded the leadership capacity in the participating schools; (3) Teachers developed leadership capacity and took on leadership roles on their teams and with their colleagues; and (4) Team members worked strategically to change instruction through a variety of approaches and targeted a variety of areas that they determined to be their school's greatest needs. The report concludes the DL project demonstrates that an intensive effort to influence and expand the leadership capacity of schools can have a positive impact on leadership practice, leadership team functioning, and support for instructional improvement. It further suggests that distributed leadership might be best thought of not as a reform in itself, but rather as a means of implementing reform. "Demographics of Distributed Leadership Schools" are appended.