Association for Public Policy Analysis & Management

Under pressure to meet educational benchmarks set by state and federal governments, school leaders have begun to rethink the ways they allocate their resources. In this context, school administrators are teaching to the test, working to align curricula and materials to the content and format of state tests (e.g., Diamond & Spillane, 2004; Firestone, Monfils & Schorr, 2004; Lyons & Algozzine, 2006).  They are also scheduling to the test by reducing time allocated for lunch, recess, and some untested subjects (Center for Education Policy, 2005; McMurrer, 2007; 2008); and even feeding to the test, increasing the calories of school lunches on days state tests are administered (Figlio & Winicki, 2002). 

Elsewhere, we have shown that school leaders are also engaged in a phenomenon we call “staffing to the test,” moving to tested grades and subjects teachers whose previous students made substantive learning gains (Cohen-Vogel, 2011). This was particularly true in lower-performing schools or schools that recently experienced a drop in performance, as measured by the grade they received from the state accountability system.

Similar findings have since been reported by Chingos and West (2011) and Grissom, Loeb & Kalogrides (2013). The former, for example, followed nearly 25,000 teachers for seven years beginning in 2001. They found that effective teachers, as measured by value-added estimates, are less likely than ineffective teachers to be reassigned to teaching positions in untested grades or subjects. Moreover, they showed that teachers in the top value-added quartile are most likely to remain in tested grades and subjects in schools that face strong pressures to improve.

What is not known, however, is how widespread ‘evidence-based’ staffing has become. By surveying all 2000+ high school principals in Florida and Texas, the authors of this paper attempt to begin to answer this question. Part of a larger study on effective high schools, the survey elicited responses to a set of items like “To what extent do you, as principal, use student achievement data to hire teachers.” 

Preliminarily, results suggest that a slight majority of principals in high schools in both states use student achievement data to make teacher hiring decisions. A larger majority in FL ad TX report using these data to assign teachers to grade levels and subjects. In both states, principals were more likely to use data in these ways when the schools they run are in some kind of correct-status under the state accountability framework.

With nearly three-quarters of principals of lower-performing high schools in both Florida and Texas reporting that they use student achievement data to move teachers between subjects and grades, implications are far-reaching. From a research perspective, scholars need to move quickly to understand the costs and benefits of evidence-based staffing for teachers and the students they serve. From a practice perspective, districts must ensure that principals are well-trained in the achievement reports they receive and how combining data over multiple years will improve teacher-level performance estimates (e.g., McCaffrey et al., 2011).