In economics, it’s commonly accepted that specialization maximizes productivity. As Adam Smith preached, specialized workers are better able to hone their skills, become more efficient, and require less transition time between tasks. When Henry Ford divided automobile production into many smaller tasks along an assembly line, for example, output improved significantly.
A recent working paper from the National Bureau of Economic Research (NBER) applies this philosophy to education, exploring whether teacher specialization (strategically assigning teachers to fewer subjects like math, reading, science, and/or social studies) improves productivity in elementary schools. Proponents of specialization argue that sorting teachers by areas of strength allows them to master subject content and spend more time on lesson planning. It may even increase teacher retention rates. In addition, some argue that early specialization might also ease the transition into middle and high school, where single-subject teaching in norm. Others caution that while specialized instructors teach less content, they teach it to a larger number of students. When specialized teachers aren’t able to get to know their students as well and tailor instruction accordingly, do learning outcomes suffer?
Author Roland Fryer explores these potential tradeoffs by randomly assigning fifty elementary schools in Houston to treatment and control groups (using a “matched-pair randomization” controlling for students’ prior math and reading scores). “Treatment” schools strategically assigned teachers to subjects in which they demonstrate effectiveness (based on value-added scores, principal observations, and/or other recommendations), while teachers in control schools continued with business as usual—where typically teachers remained with the same set of students all day.
The results were a surprise: Fryer finds that, rather than improving academic outcomes, two years of teacher specialization negatively impacted student achievement in both reading and math. On average, specialized teachers were about 6 percent less effective as their non-specialized peers (as measured by student scores on state tests in math and reading). (Results were even worse for special education students and students with more inexperienced teachers). Students in treatment schools were also suspended at slightly higher rates and had marginally worse attendance.
Fryer speculates that the benefits of sorting teachers based on comparative advantage might be negated by “inefficient pedagogy due to having fewer interactions with each students,” more transition time required between classes, and other coordination-related costs. But most important, he argues, is the fact that specialized teachers have more students and are less able to modify individual instruction effectively.
Before we write off specialization entirely, however, we should note that the study has several important limitations. It was only conducted in one large, urban district, so results aren’t applicable to elementary school students in all districts and states. Participating students were predominantly minorities and had lower achievement scores than other students in the Houston Independent School District. And as Fryer himself explains, schools were only able to reallocate teaching assignments within any given grade at their school; many were limited by the number of staff members teaching certain grades.
Although these findings suggest that the benefits of teacher specialization in elementary schools may not outweigh the loss of individualized instruction, more research should be done to better understand the tradeoffs.
SOURCE: Roland G. Fryer, Jr, “The 'Pupil' Factory: Specialization and the Production of Human Capital in Schools,” National Bureau of Economic Research (April 2016).
