Spreading the Wealth

Performance-based accountability (PBA) has provided educational leaders with incentives to use achievement data to plan for school improvement. In fact, there is evidence that they are using test score data for decisions about everything from the curriculum to what is served for lunch. In the article, “Staffing to the Test,” we previously documented that staffing too is data-driven, with administrators moving to tested grades and subjects teachers whose students make substantive learning gains. But, what are the implications of PBA for the assignment of students?

PBA is, in large part, intended to help concentrate attention on students who struggle academically.  As such, we might expect schools to respond to accountability pressures by assigning lower performing students to better teachers, for example.  Specifically, we might anticipate school administrators to target their instructional resources, including teachers whose previous students demonstrated the highest achievement gains, on students whose performance places them below the state’s proficiency cut-off and, in particular, on “bubble” kids who have the highest statistical probability of moving into the “proficient” category. Because lower performing schools are also more likely than their better performing counterparts to feel the threat of sanctions (e.g., reconstitution; school closure), we further predicted that educators in these schools would be more likely to use test scores in student assignment and to assign lower performing or on-the-bubble students to better teachers.

In our case studies of elementary schools in five Florida districts, we interviewed principals, assistant principals, and teachers to understand how student assignments are made.  In 1999, as part of its PBA policy, Florida implemented a system through which schools received a grade of A through F, with students’ scores on the Florida Comprehensive Achievement Test (or FCAT) weighing heavily into the grade calculation.  In each district, we used school grades to select one high- and one low-performing school to examine whether a school’s performance conditioned the ways in which students are assigned.

In nine of the ten schools studied, participants reported that performance data played a central role in the student assignment process.  Performance data included students’ scores on standardized tests (the SAT-10 in early elementary and FCAT in upper elementary) and course grades.  Due to the near ubiquity of the practice of using performance data in student assignment decisions, we saw little to suggest that schools with different grades on the state’s accountability system differed in their propensity to make data-driven assignment decisions.

Even as they reported using performance data in the assignment process, school leaders in high- and low-performing schools alike appeared to be working to achieve balance above any other objective as they assigned students to classes and teachers. With one exception, they did not speak of efforts to pair students with teachers who had previously demonstrated success with similar students.  Nor did they report assigning “bubble kids” to their top-performing teachers.  In fact, educators in the study used student performance data to populate classrooms that intentionally spread high, middle, and low performers equally across grade-level teachers.  Our evidence suggests that principals worked to achieve balance because it is, in their opinions, not only “fair” but also because it helps provide for well-managed classroom environments conducive to learning.

From a policy perspective, while PBA may have encouraged principals of elementary schools in Florida to use performance data in student assignment decisions, it has not lead them to do so in ways that redistribute their teacher workforce. Principals do not work to ensure that their lowest-performing students are assigned to the most effective teachers, instead preferring to promote a sense of fairness among their staffs and, in one teacher’s words, to “spread the wealth.” 

The full study can be found in: Osborne-Lampkin, L. & Cohen-Vogel, L. (2014). “Spreading the wealth”:  How principals use performance data to populate classrooms. Leadership and Policy in Schools, 13, 188–208.