Behind the walls of AIM Academy, a Philadelphia independent school for bright children with learning differences, a unique leadership model is taking shape. The leadership flows from the top down, the bottom up, and from side to side.

This multidirectional model allows for teacher leaders within the school to organically emerge and simultaneously impact the broader school while remaining in the classroom. This equals significant outcomes for students with unique learning profiles coupled with professional growth for teachers interested in exploring leadership without having them step away from the critical instruction they deliver each day.

Distributed leadership models are not new or specific to our school. Research in the late '90s led to the first formal models being established in schools, though one could argue that communities of practice and collaborative leadership are notions that have been around for thousands of years.

In schools, though, this continues to be an emerging practice with a variety of models — some of which have been effective in promoting outcomes for students and some that have fallen short of that intended goal. All of the successful models have three things in common:

• They provide collaborative and shared leadership opportunities for faculty members already in the school.
• The distributed leadership team members address real needs and meet the goals of the school.
• Effective leadership teams are given the ability to lead.

In other words, these teams are not just delegated tasks; they are given real projects and ownership over specific and important initiatives (McREL 2005).

The model at our school is loosely based off the approach developed by the Center for Educational Leadership at the University of Pennsylvania in conjunction with the Annenberg Foundation. This model, based on the notion of collaborative leadership, has resulted in significant improvement in student outcomes in the hundreds of schools (Bass 2007).

Penn's model was designed for larger institutions and has been implemented successfully in districts where the needs are quite different than at a small independent school like ours. Though broad goals of the leadership model and some aspects of implementation might look different at AIM than they would say in the School District of Philadelphia, one could argue the outcomes are consistent across environments.

An effective distributed leadership program decreases faculty attrition and improves faculty perceptions of the school as measured by direct teacher survey data (Spillane 2001). Faculty satisfaction, as measured by teacher recruitment and retention, also increases (Riordan 2003).

Naturally, teacher quality is the greatest determining factor on student achievement, but one would be remiss to overlook the logical assumption that highly qualified and highly satisfied teachers have an even greater impact on student outcomes.

Building leadership within the faculty is healthy for the whole school community both in and out of the classroom. When effective leadership teams are in place, teacher engagement for those both on and off of the leadership team improves. Teachers report increases in participation in school events, higher morale and a stronger understanding and commitment to the culture of the schools (Harris 2000).

It is clear that teams matter, and leadership models that rely on individuals are less effective than those that rely on collaborative teams (Harris 2007, 2008). Across the literature, schools that employ leadership teams see more positive student outcomes than those with stand-alone leaders (Smylie, Conley & Marks 2002), (Marzano 2007), (Waters & McNulty 2005).

It is from this critical thread in the literature that we began exploring and then expanding our distributed leadership program — the Leadership for the Future Collaborative.

As one starts digging into the literature, it is quickly apparent that teams alone will not result in significant gains for students. This is because the quality of leadership counts.

Effective teams are the recipients of formal training, and training is actually a key determinant factor of the success of any distributed leadership model (Townsend 2007). Therefore, training is a key component of our Leadership for the Future Collaborative with both formal and more discussion-based cohort style sessions.

Teachers-turned-leaders create instructional issues for schools in that many skilled teachers leave the classroom to pursue leadership. Our program attempts to circumvent this issue by creating leadership projects that do not require the team members to be out of the classroom for significant portions of the day.

While instruction is clearly at the heart of all classrooms, the training our teachers receive is specific to the type of students enrolled. Teacher trainings include a number of research-based programs for average to above children with language-based learning disabilities such as dyslexia and dysgraphia.

It takes years to acquire the instructional and pedagogical knowledge base of an AIM teacher. Therefore, coming out of the classroom to lead presents greater concerns for student outcomes here as opposed to an institution where the instruction is less specialized.

An important piece of our leadership program is that it actually creates a more sustainable model for schools with highly specialized instruction like ours. Many of the schools initially studied by the Penn Center for Educational Leadership through the $4.9 million grant from Annenberg Foundation included underperforming children.

In these challenging demographics, collaborative distributed leadership teams were seen as key in the success of the school and its students. Student achievement was rising in all of the nearly 200 schools studied (Hallinger & Heck 2008).

Another key to success is the presence of a school's chief instructional leader, the principal. Distributed leadership teams resulted in principals who spend more time in classrooms, and a principal's presence has been linked to improved student outcomes (Day, Gronn & Salas 2006).

Our program involves a small team selected each year to work on specific projects that address both personally and institutionally inspired goals. Therefore, the work is both personally rewarding as well as important to the strategic goals of the school.

The team meets for a summer leadership institute where they learn about leadership topics ranging from communication, to culture, to change models, to school finance, etc. In the fall, as the rest of the faculty returns, the team and their projects are officially announced, and authority for the team members is publicly sanctioned.

As the projects evolve, a series of discussion-based meetings about critical topics occur. These meetings, which happen every six to eight weeks, focus on sharing progress, challenges and relevant literature.

Members of the team have described the experienced as "an excellent exercise in reflective practice" and "an opportunity to troubleshoot real-time issues while balancing the theoretical simultaneously." Because the work is useful, tangible and real, "the projects had an immediate effect on how to ameliorate problems and develop solutions in the school."

In the words of one of our first-year cohort members, "The experience allowed me to hone my leadership philosophy while maintaining a grounded approach to moving the school forward in a productive manner — it enabled me to meet real goals for the school."

With such success and so much research to support this model for faculty leader development, we intend to strengthen and expand this powerful and effective program.