The Effect of Co-Teaching on Student Performance

• In Event: Effectiveness and Equity in Education for Students With Disabilities

Abstract

Objective

Federal legislation requires that students with disabilities be educated in the least restrictive environment, resulting in the majority of students with disabilities (SWDs) being educated in the general education classroom for most of the school day (USDOE, 2015). Co-teaching – in which general educators and special educators work in the same classroom – has emerged as a strategy for meeting this goal. A substantial body of literature has been dedicated to co-teaching, with the majority of this work offering guidance for implementation (e.g., Villa et al., 2013; Murawski & Dieker, 2004). However, despite its prevalence and acceptance as practice, there is virtually no empirical research estimating the effect of co-teaching on student outcomes (Solis, Vaughn, Swanson, & McCulley, 2012).

Framework

The limited evidence we have suggests that co-teaching arrangements are implemented in ineffective ways. Several qualitative studies have suggested that general and special education teachers in co-taught classrooms do not regularly engage in the types of instructional practices known to meet the needs of SWDs. The the goal of this paper is therefore to produce the first plausibly causal estimate for the effect of attending co-teaching classrooms on measured student educational outcomes.

Methods and Data Sources

We utilize student-level longitudinal administrative data from the Massachusetts Department of Elementary and Secondary Education (DESE). We derive the final dataset for estimation by matching DESE’s longitudinal student and teacher datasets for each year from 2008 to 2018.

We apply a fixed-effects approach to measure the causal effect of a attending a co-taught classroom on student standardized math and ELA test scores. The coefficients of interest are estimated using variation in the performance of students who are observed in more than one classroom setting. We employ two definitions of co-teaching derived from our data – one broad including all classrooms with two assigned teachers and one narrow where one teacher is labeled as co-teacher.

Results

We find evidence that attending a co-taught classroom leads to lower performance among SWDs. Relative to attending a classroom led by a single teacher, attending a classroom with multiple teachers where at least one of them is classified as a co-teacher (our narrow definition) has either no effect or a modest negative effect on the test scores of students without disabilities in both math and ELA. However, attending a co-taught classroom according to our broader definition has a significant negative effect on the ELA scores of students without disabilities. We find no significant effect in math from a student without disabilities attending a co-taught classroom according to the broad definition.

Significance

The use of co-teaching as a model for educating students with disabilities in the general education classroom has exponentially grown. Our study provides evidence that it does not lead to improved outcomes for students. When coupled with studies that directly observe the quality of classroom practices in co-teaching (see Wexler et al., 2018 as one example), studies such as ours will contribute to an evidentiary base that can support leaders in making decisions that are informed by research.

Authors

• Nathan Jones, Boston University

• Marcus Winters, Boston University