A new report examines whether the effect of a teacher’s value added persists over one to two years and across subject areas. In particular, it asks if the impact of an English language arts teacher has any bearing on a student’s math achievement. Prior research has provided some evidence that ELA instructional effects may be generalizable across subjects, given the applicability and transferability of reading and language skills.
Using data from the New York City and Miami-Dade school districts, Benjamin Master and colleagues use student records in grades four through eight, where current and prior year achievement data are available for students. They investigate the persistence of teachers’ value added effects on student achievement in the first and second year after they teach a student, distinguishing between short-term, test-specific knowledge and longer-term, generic knowledge that accumulates. The methods are complex: They attempt to isolate teacher-specific value added that persists both in the same subject and into another subject (either ELA or math), and to isolate this from teacher spillover effects that stem from same-year instructional collaboration with peers—while also estimating how much typical “decay” one might expect of student’s prior long-term knowledge—and also controlling for various student, school, and classroom characteristics.
The results are thankfully easier to follow than the methods. First, most of the previously assessed long term knowledge within the same subject area that students learned in the prior two years persisted into a third year. Second, 26–29 percent of a teachers’ within-subject value added on student learning in New York City persisted into the subsequent school year, with somewhat higher percentages in Miami-Dade. Third, however, ELA teacher effects persisted at a much higher rate for students across both subjects one and two years later. For example, in New York City, the two-year cross-subject ELA persistence rate is approximately 42 percent of the within-subject two-year persistence rate estimate; whereas the comparable figure for a math teachers’ cross-subject persistence rate is less than 1 percent, which is not that surprising, since we don’t expect math learning to contribute to ELA learning.
The authors summarize their study this way: “Learning due to ELA instruction appears to impart long-term knowledge and skills that are reflected not only in short-term ELA scores but also in future test scores in both subject areas.”
They also caution that ELA teachers’ contributions are diffuse and a large portion of their instructional impact in current value-added models may be going undetected or ascribed to other teachers in other subject areas or years. Their solution is to pay more attention to team-level measures of value-added or models that simultaneously account for multiple teachers’ contributions in multiple subjects. We’re all for precision, but talk about making value-added calculations even more complicated and hard for educators—and the rest of us—to understand!
SOURCE: Benjamin Master et al., “More Than Content: The Persistent Cross-Subject Effects of English Language Arts Teachers’ Instruction,” Educational Evaluation and Policy Analysis (February 2017).
