Every family sends their child off to school, hoping the teacher will effectively meet their child's unique needs. In trying to assess teacher effectiveness, however, many states and districts assume that teachers are equally effective with different types of students. A recent paper by W. Jesse Wood and colleagues from CALDER examines this assumption, exploring whether some teachers may be more effective with students with disabilities as compared to their effectiveness with students without disabilities.

The researchers used math and English language arts (ELA) scores from the Los Angeles Unified School District over a 10-year period (2007-08 to 2017-18) to create two value-added measures for each teacher—one that measured the teacher's contribution to the achievement of students with disabilities and one that measured that same teacher's contribution to the achievement of students without disabilities. The researchers then compared the teacher's rankings in each measure to their teacher peers across the district, looking at effectiveness with students with disabilities relative to effectiveness with students without disabilities.

The study found that teachers who are effective with one group of students also tend to be effective with another group. However, some teachers have a relative advantage for teaching students with disabilities, while other teachers have a relative advantage for teaching students without disabilities. In other words, a teacher may be effective overall—as measured by value-added data—but is comparatively more effective in driving academic growth with one student group than with another in relation to how other teachers perform with that group.

In looking at student assignment patterns, researchers found that students with disabilities tend to be assigned to teachers who are less effective overall. However, these students also tend to be placed with teachers who—while less effective overall—are comparatively more effective in teaching students with disabilities than in teaching students without disabilities. This assignment pattern holds true for ELA at the district level, and for both ELA and math at the school level. As it is unlikely that most school leaders look at disaggregated, value-added data when determining teacher placements, this finding suggests that leaders may be relying on other information to identify and leverage teachers with a particular strength in working with a specific group of students.

While it is promising that schools are assigning students with disabilities to teachers who are comparatively more effective in working with this student group, it is concerning that these students are consistently taught by less effective teachers overall. Students with disabilities are often one of the lowest-performing subgroups on assessments, and it is extremely difficult for students to make achievement gains if they disproportionately lack access to effective teachers.

This study also suggests there could be benefits to looking at teachers' value-added data in a more nuanced way. In many states and districts that use value-added data, all students taught by a teacher are grouped together regardless of different student characteristics. The findings of this study suggest that disaggregating value-added measures for different students, such as students with disabilities, could provide more useful data about teachers' specific strengths and areas for improvement in advancing the achievement of different students. Some states, including Delaware and Massachusetts, already break down teachers' value-added data in this way; other states could move in this direction to increase the usefulness of value-added data for both understanding teachers' effectiveness and informing teacher assignment at the school level.