New study examines 'strategic retention' of teachers by effective principals
Numerous studies have linked principal effectiveness to overall reduced teacher turnover. These studies, however, have not differentiated between turnover of high-performing and low-performing teachers. While lower average teacher turnover is associated with better student outcomes, turnover of lower performing teachers may be a good thing for a school and for students.
A study published online today in the American Educational Research Journal, a journal of the American Educational Research Association, backs up prior research indicating that highly rated principals—as determined by assessments from their supervisors and the teachers in their schools—see lower average rates of teacher turnover.
However, the new study, by two Vanderbilt University researchers, suggests that this lower turnover is concentrated among teachers with higher scores on classroom observation measures and higher student test-performance growth scores (also known as value-added scores).
Looking at data from Tennessee, a state that collects multiple measures of principal and teacher performance, Vanderbilt's Jason A. Grissom, an associate professor of public policy and education, and Brendan Bartanen, a doctoral candidate, find that highly rated principals use "strategic retention," retaining high-performing teachers while moving out lower performers.
Notably, these principals appear to focus on just one performance measure—teacher observation scores—to identify low-performing teachers to move out.
"We find that under highly rated principals, teachers who score higher on multiple performance measures are more likely to stay than under lower rated principals," said Grissom. "By contrast, teachers who receive the lowest classroom observation scores leave at substantially higher rates under an effective principal, regardless of whether they have high or low value-added scores."
"We suspect this is because principals collect observation data themselves throughout the year and use these observations to learn about performance," said Grissom.
In contrast, Grissom noted, in Tennessee, as in many states, value-added scores are not returned to principals until fall, meaning principals cannot take them into account in making retention decisions for the next year.
"So much of the rhetoric around teacher evaluation focuses on student test score growth, and principals around the country are held accountable for test score growth in their schools, so we were surprised at first by this finding," Grissom. "But in the context of other research on the timing of the return of teacher value-added scores and principals' questioning of what they learn from those metrics, it makes sense that principals would focus on observation scores in making decisions about teachers."
"States have invested a ton in the last decade in large-scale 'multiple-measures' teacher evaluation systems," said Grissom. "If states and districts want principals to make use of student test score growth information to remove ineffective teachers, they need to recognize that even effective principals are not able to do so if they are not given access to this information in time."
Grissom and Bartanen find that effective principals likely use informal means, such as "counseling out," to remove low-performing teachers, rather than relying on administrative procedures. They also find that these teachers typically exit teaching rather than move to another school in the same district.
"It's clear that school leadership is important if a school community wants to keep its best teachers," Grissom said. "And if teacher evaluation systems are supposed to help transform the teaching workforce, strong school leadership is essential."
"Our findings also speak to the importance of achieving an even better understanding of principals' talent management roles, especially their specific strategic behaviors," said Grissom. "Principals are an important and largely ignored component of linking large-scale teacher evaluation systems to the reshaping of the teacher workforce. States haven't focused on this issue nearly enough."