Editor's note: This is the sixth and final post in our latest blog series by John Chubb, "Building a Better Leader: Lessons from New Principal Leadership Development Programs." See here, here, here, here, and here for prior posts.
This week I summarize what policymakers can learn from alternative leadership development models—and how these programs, and others like them, can be improved upon.
1. States should measure the added value of school principals and of leadership development programs. For all of the commonalities among the exemplar programs examined here, the evidence is a long way from definitive. It is merely the best that can be gleaned from the data now available. Leadership programs could—and should—do a much better job of tracking their graduates to improve their own offerings. But a proper analysis of principal effectiveness requires achievement data and background information on individual students and teachers in the schools that new principals lead. The states make the rules for what data school districts report and what indicators are derived from those data. Leadership programs cannot estimate the effectiveness of their graduates without state cooperation.
Policymakers should therefore require state departments of education to begin estimating the added value that principals bring to their schools. Policymakers should also require public school principals to report where they received their certification and training. This would allow states to estimate not only the effectiveness of each principal, but the effectiveness of the institution or program that trained them. States are already making these calculations for teachers and their training programs. They should do the same for school leadership. To be sure, value-added measures for teachers are not perfect. They have reliability issues, and in any case, they should not be more than one of several measures of teacher effectiveness. The same with the value-added measures for principals and leadership development programs.
2. Schools, school systems, and school management organizations should recruit prospective leaders from their own ranks, beginning with teachers, and develop them systematically over time, consistent with research-tested standards. Today’s methods of leadership development are notoriously passive and non-selective. Most school leaders decide for themselves that they have the right stuff and then seek training in colleges and universities without rigorous requirements for admission. Schools and school systems need not be limited by this age-old process. They could and should proactively identify prospects themselves. The exemplar programs highlighted in this series regard selection as critical—not just anyone is likely to become a highly effective leader.
Data and research could take some of the guesswork out of the evaluation of leadership potential. If states provide value-added estimates of principal effectiveness, researchers will be equipped to evaluate possible causes of that effectiveness—intelligence, prior success in the classroom, teaching or administrative experience, personality type, and so forth. Schools will then be better able to judge who to advance for leadership training and promotion.
3. Schools should organize themselves explicitly for the internal development of leadership skills. The exemplar programs view leadership development as a lengthy process of instruction, practice, feedback, and more practice—best obtained on the job. Schools and school systems or management organizations should rely less on external leadership development programs and more on development internally. Just as great schools create the conditions for rookie teachers to grow into proficient pros, schools should do the same for budding leaders. They should create scaffolded opportunities for educators to lead, from teachers leading small teams of teachers to principals working with multiple departments (not yet on a school-wide scale). As important, schools should provide systematic support in the form of coaching, evaluation, and formal instruction in leadership skills, If schools paid more attention to leadership development themselves, they would have to pay less to external programs to develop leaders for them.
4. Research should work to identify high-impact leadership behaviors. Extant research into effective school leadership is crude at best, identifying correlates of successful leadership with few statistical controls to provide confidence in cause and effect. Leadership programs are invariably guided by competency frameworks of some kind, almost always unsupported by rigorous evidence for their various elements. The exemplars have frameworks as well. In practice, the exemplars place the greatest emphasis on people skills, personal management, team building, teacher development, and the creation of a strong school culture. The emphasis on these soft skills is evidence that top programs believe that the most important thing that principals do is fill schools with committed teams of top teachers: great leaders, great teachers. It would help future leaders enormously if the behaviors of successful leaders could be specified more explicitly. We now have the benefit of empirically validated instructional frameworks to guide the development of effective teachers. The ambitious MET study recently showed that several models of good teaching—such as Danielson’s “Framework for Teaching”—are associated with greater student achievement. Research might provide the same for school leaders.
5. Leadership programs themselves should live or die by research—or at least be better disciplined by it. Schools and school systems will not meet the needs of leadership development alone. They will not conduct the research and they will not acquire requisite expertise. Colleges and universities, not-for-profits, for-profits, and various scale organizations will inevitably have roles to play. If the effectiveness of their programs could be measured by state value-added assessments—providing one of perhaps several objective measures of effectiveness—programs would have incentive to improve. Effective programs will become known to prospective leaders and to the employers who eventually hire them. Strong programs will attract more students and place more graduates. Weak programs will find ways to upgrade their programs or face questions from students, employers and policymakers about their ineffectiveness. Most important, programs will be encouraged to use research and experimentation to determine what leadership requires. Our guess is that the answers lie in programs that offer training from the beginning of the leadership pipeline to the end, beginning with classroom teachers; that provide intensive, well-supervised residencies; that offer strong support beyond graduation; and that emphasize the soft skills of leading people. In the end, leadership development will not improve until schools and policymakers insist on evidence that they make a difference for students. Educational leadership is ultimately about giving students more opportunity to learn. Student learning should be our arbiter of how great leaders are developed.
John Chubb is the president of the National Association of Independent Schools.