Although principals have long been seen as important to the success of schools, a new set of working papers from some prominent education researchers aims to promote a better understanding of the extent to which school leaders matter and why.
For instance, one study of the 345,000-student Miami-Dade County, Fla., school system finds that the most effective principals appear to be particularly adept at weeding out weak teachers and keeping strong ones. Another paper looking across Texas concludes that the skill of a principal is most important to student outcomes in the most challenged academic environments: schools serving large numbers of low-achieving students living in poverty.
Some of the research also analyzes the distribution of principals. One paper, for example, suggests that schools with the most disadvantaged populations attract principals who have less education and experience, and who attended less selective colleges, compared with leaders of better-off schools.
The working papers are scheduled to be discussed this week at a Washington conference sponsored by the National Center for Analysis of Longitudinal Data in Education Research, or , a federally funded center housed at the Urban Institute, a Washington think tank.
鈥淓verybody says principals make a difference, but there鈥檚 really been no systematic effort to try to estimate the extent to which they make a difference, how they make a difference, and how they鈥檙e distributed across schools,鈥� said Jane Hannaway, who is the director of the Urban Institute鈥檚 education policy center and also heads CALDER, which is publishing the working papers. 鈥淲hat these papers do is try to open up that box, the principal box, ... but this is hardly the final word. Stay tuned.鈥�
Among the scholars scheduled to present papers at the conference are Eric A. Hanushek, a senior fellow with the Hoover Institution at Stanford University; Helen F. Ladd, a professor of public policy and economics at Duke University; and Susanna Loeb, the director of the Institute for Research on Education Policy & Practice at Stanford.
鈥楧etective Work鈥�
Mr. Hanushek, who co-wrote a study titled 鈥淓stimating Principal Effectiveness,鈥� notes that it鈥檚 particularly difficult to determine the influence of principals on student achievement, as distinguished from the influence of the many other factors in schools. His study involved working with the Texas Education Agency to combine different state data sets to devise a principal value-added measure.
鈥淚t takes some detective work to pull out the systematic components,鈥� he said.
The results suggest that principals do matter, especially in high-poverty schools.
鈥淚n the simplest version, a good principal has a big impact on high-poverty schools, and bigger than in low-poverty schools,鈥� Mr. Hanushek said. 鈥淪o if you look at the range, you see that principals seem to be spreading out the performance distribution in [those] schools.鈥�
At the same time, he said, his research so far has not been able to explain the particular reasons the principal seems to matter.
Another working paper to be presented this week, 鈥淲hat Makes an Effective Principal? The Characteristics and Skills of Quality School Leaders,鈥� begins to answer that question through an analysis of theMiami-Dade County district.
It finds that the principals who are effective in improving student achievement tend to have a higher turnover rate among their teachers, but that鈥檚 because those actions are producing a stronger workforce.
鈥淲e see big differences across schools, and across principals in the same school, in their ability to attract effective teachers and get rid of ineffective teachers,鈥� said Ms. Loeb from Stanford, a co-author of the study.
The paper also finds that principals become more effective as they acquire more experience overall, and as they gain greater experience at a particular school.
Ms. Loeb also co-wrote another working paper to be presented at the CALDER conference that examines the relationship between the time principals spend on different types of activities and school outcomes, including student achievement, teacher and parent assessments of the school, and teacher satisfaction.
That study, 鈥淧rincipal Time-Use and School Effectiveness,鈥� highlights the importance of a principal鈥檚 skill in 鈥渙rganizational management,鈥� such as hiring and managing staff members, implementing professional development, and managing budgets.
鈥淭here is a push right now for instructional leadership, and I like that as long as you think of it more broadly,鈥� Ms. Loeb said, 鈥渂ecause in some cases it鈥檚 going in and observing in the classroom. That takes a lot of time, and it doesn鈥檛 seem to be effective, on average.鈥�
She added: 鈥淲e鈥檙e not saying you don鈥檛 want to do any of that,鈥� but the principal will likely see far greater benefit from spending more time on organizational-management tasks.
Experience Matters
A working paper co-written by researchers at the University of Florida, the Santa Monica, Calif.-based RAND Corp., and Columbia University鈥檚 business school, meanwhile, examines the characteristics of school principals as they relate to student performance on standardized tests in New York City.
Titled 鈥淪chool Principals and School Performance,鈥� the study finds 鈥渓ittle evidence鈥� that a principal鈥檚 own education or work experience before becoming a principal has an impact on school performance. It does find 鈥渟ome evidence that experience as an assistant principal at the principal鈥檚 current school is associated with higher performance among inexperienced principals.鈥�
At the same time, the researchers find a positive relationship between principal experience and school performance, particularly for math test scores, as well as for having fewer student absences.
鈥淥ur clearest finding is that schools perform better when they are led by experienced principals,鈥漷he working paper says. 鈥淭he experience profile is especially steep over the first few years.鈥�
The researchers suggest that their conclusion, even while according with 鈥渃ommon sense,鈥� has important policy ramifications.
For one, it alerts district administrators to the potential costs of having experienced principals leave their jobs, and, on the flip side, the benefits of keeping them in place.
At the same time, it says, the 鈥渢endency for less advantaged schools to be run by less experienced principals could exacerbate educational inequality.鈥�
Still another paper Ms. Loeb co-wrote looks at the distribution of school principals. Among its findings is that schools serving large proportions of low-achieving minority students from low-income families tend to have principals with less education and experience.
鈥淭hese students are more likely to attend a school that has a first-year principal, a principal with less average experience, a temporary or interim principal, a principal without a master鈥檚 degree, and a principal that went to a less selective college as compared to their more advantaged counterparts,鈥� the paper says.
The researchers point out that the finding mirrors other research showing a similar distribution of teachers, and of teacher turnover at such schools.
鈥淭he similarity between the sorting of principals and teachers islikely to not be coincidental, but driven by a shared preference for schools serving less at-risk populations,鈥� says the report, 鈥淧rincipal Preferences and the Unequal Distribution of Principals Across Schools.鈥�
The study Mr. Hanushek co-wrote arrives at a similar conclusion about the general distribution of school leaders in Texas.
鈥淧rincipals seem to behave much like teachers,鈥� he said. 鈥淭hey seek out schools with good, high-performing kids and not too many disadvantaged kids.鈥�
