The Department of Education’s planned release of the first year of School Improvement Grant (SIG) data has important implications for the future of the program – and for reform generally. While much of the early attention devoted to SIG has focused on its cost-effectiveness, in light of the ongoing debate over school and district human-capital policies the question of which of the four SIG models is most effective deserves greater attention. As Education Sector documented in its original Portrait of School Improvement Grantees, more than 70 percent of SIG schools chose the “transformation” model of school improvement, which requires replacing the principal and implementing a number of data management and teacher evaluation related reforms. Another 20 percent of schools chose the “turnaround” model, which is similar to transformation except that it also requires that a school replace at least 50 percent of its teachers. The contrast between these models means they will inevitably be linked to the larger question: Can schools and districts fire their way to higher student achievement?
Though it rarely says so publicly, the reform movement is deeply invested in the notion that they can. A wealth of research has demonstrated that teacher quality is the greatest in-school determinant of student achievement, so replacing poor teachers with average or above average teachers should lead to higher levels of achievement (assuming that suitable replacements can really be found). Theoretical explorations of this idea have produced some striking results. For example, one paper suggested that the optimal replacement rate for teachers could be as high as 80 percent a year, when the negative effects of putting more inexperienced teachers in the classroom were balanced against the benefits of greater selection.
In the real world, of course, adopting such an approach is both unfeasible and undesirable. At least one study has found that high levels of teacher turnover are detrimental to student learning. Moreover, since we need at least 3 million teachers in our classrooms every year, attempting to improve teacher quality by replacing 50 or 80 percent of teachers a year would require an improbably huge talent pool from which to draw – especially in districts that struggle with teacher recruitment and retention. For these reasons (and many others) even the most aggressive teacher selection policies have led to the removal of at most 5-6 percent of teachers a year. Unfortunately, while some reformers have argued that this should be enough to raise student achievement, in practice it has proved extremely difficult to gauge the impact of such policies. After all, even at the district level test scores can fluctuate for many reasons.
Given this uncertainty, it will be tempting for both sides of the teacher effectiveness debate to over-interpret the results from SIG. But no matter what the data ultimately reveal, it will be important to remember that there are major differences between school and district-level turnaround. For example, it is possible that the effects of the turnaround model are driven by purely school-level dynamics related to teacher selection that do not operate at the district level. Similarly, turnaround schools may still be relying on “last-in-first-out” policies – (In which case, why choose the turnaround model to begin with?) – or they may be misidentifying their low-performing teachers for other reasons.
Despite these caveats, in the absence of natural experiments at the district level, SIG is likely to provide us with important information about the real-world effects of replacing significant numbers of teachers in a short period of time. So far, the only study to examine the effects of SIG provides tentative support for the turnaround model and for the notion that removing low-performers from the teaching workforce can lead to higher student achievement – but this finding has yet to be replicated. Let the debate begin!
Written by Education Sector Policy Intern David Dickey-Griffith
Photo Credit: NBC North Dakota News