As a society, we seem to agree that good teaching is an important lever by which we can improve education—maybe the most important. Organizations like Teach for America endeavor to recruit teachers from elite schools with the idea that these teachers will raise student achievement. And much of the political discussion regarding our schools concerns how we can hold teachers more accountable—whether through merit-based pay or the elimination of tenure.

Indeed, there seems to be growing consensus around what Kim Marshall argues in Rethinking Teacher Supervision: “the quality of instruction is the single most important factor in student achievement.” But if great teaching is so important to the success of our schools, shouldn’t we also be focused on helping teachers improve their instruction? And, to this end, shouldn’t principals be asked to spend much of their time doing just that?

The request that principals serve as instructional leaders of their schools is actually relatively new. According to Ed Finkel in Principals as Instructional Leaders, principals have historically been managers, “making sure every student has a desk, the buses are on time, and cafeterias are supervised.” Today, because of budget and personnel cuts, and increased accountability, principals must also take on responsibility for improving instruction in schools.

Many feel this shift toward principals as instructional leaders is a positive change. Paul Bambrick-Santoyo, managing director of seven high achieving charter schools in New Jersey, argues in his book Leverage Leadership that being an exceptional school leader means being focused on instruction above all else. “What really makes education effective,” he insists, “is well-leveraged leadership that ensures great teaching to guarantee great learning.”

Yet Bambrick-Santoyo also asserts that this critical work is largely absent from what most principals do on a daily basis. When researchers from Stanford and Vanderbilt followed 65 principals in Miami-Dade’s public schools to see how they were spending their time, they found that principals spent less than six percent of their time improving instruction, via coaching, evaluating, and observing teachers or implementing professional development specifically concerning instruction.

Interestingly, the study also found that the lowest performing schools had principals who spent more time on administrative tasks, such as fulfilling compliance requirements, and managing discipline, schedules, attendance, and standardized testing. Whether a school’s poor performance causes a principal to spend more time on administrative tasks or vice versa remains an open question—the study merely observes correlation, not causation.

The LAUSD principals I spoke with reported devoting more time—an average of 18 percent of their week—to instructional leadership. Still, they said this was much less time than they would like it to be. They cited district meetings, compliance obligations and documentation, and audits as obstacles that prevented them from spending more time specifically addressing instruction.

In order to devote more time to supporting teacher instruction, the LAUSD principals I spoke with articulated a need for more support personnel to take on some of the many roles principals are currently performing themselves—such as operations managers, counselors, or disciplinarians. They also universally agreed they would simply need more hours in the day.

One model being experimented with in Kentucky is to hire a School Administration Manager, or SAM. The SAM is usually a staff member who takes on some of the principal’s most time-consuming management tasks such as office work and student discipline. This, in turn, frees up principals to lead instruction. In schools with SAMs, the Wallace Foundation found that principals increased the time they spent on instructional leadership from 36 percent to over 50 percent of their day.

Additionally, the way this instructional support time is spent matters a great deal. The same study done by Stanford and Vanderbilt found that “walk-throughs”—which are the most common instruction activity principals do—were actually negatively correlated with student achievement. On the other hand, coaching was associated with gains in student performance.

Principals also need high quality professional development that helps them to effectively coach their teachers. Currently less than four percent of Title II federal funding is spent on principal training. Both the National Association of Elementary School Principals (NAESP) and the National Association of Secondary School Principals (NASSP) have recommended that figure be increased to ten percent.

In the end, if we hope to improve instruction, we need to redefine the roles of school leaders. This likely requires removing or shifting some of the many administrative tasks that consume principals’ days, and encouraging and training principals to do the important work of leading instruction.