TCLD Page Banner

Stockard, J., Wood, T. W., Coughlin, C., & Khoury, C. R. (2018). The effectiveness of Direct Instruction curricula: A meta-analysis of a half century of research. Review of Educational Research, 88(4), 479–507. doi:10.3102/0034654317751919

Summary by Dr. Sharon Vaughn


Effective instruction for students with learning disabilities is explicit and systematic with a highly sequenced and specified set of tasks that build students’ learning in critical academic areas such as reading and math. Direct Instruction (DI) is often referred to as representing this type of effective instruction, particularly for students with learning difficulties. Findings from studies examining the numerous curricula that are derived from the DI work of Siegfried Engelmann and his collaborators (Engelmann & Colvin, 2006) are reviewed in this synthesis. DI is based on the premise that all students can learn through (a) unambiguous instruction, (b) appropriate examples aligned to the instruction, (c) sequenced instruction that supports learning goals, (d) strategic work toward mastery of new skills so they can be used effortlessly, and (e) appropriate placement in instructional programming so learning is fun and accelerates at a rapid rate.

The purpose of this review of DI is to address limitations of previous reviews by expanding the focus of studies included so that the review would be as comprehensive as possible. The authors also sought to determine whether estimates of findings varied across subjects, publication outlets, methodological approaches, participant characteristics, and intervention procedures.


As this paper is a review of existing literature rather than an independent study, the method involved describing the studies that had been previously conducted by a large number of scholars over a 60-year period—since DI was first developed. After an extensive search of all studies, 328 studies (including dissertations) involving almost 4,000 effect sizes were deemed acceptable for analysis. These studies, which included a range of participants, settings, and measures, were reviewed and the key descriptive information and findings from each study were coded using a specified code sheet. Variables related to the participants, the intervention, the measures, and other context-related characteristics were reported on the code sheet so that they could be analyzed in the findings. The review examined effects of DI on academic areas such as math, reading, language, and spelling.

Key Findings

  • DI was consistently associated with moderate to large impact (effect sizes). In other words, when DI approaches (or curricula) were used in the study, the participants involved in the DI approaches performed, on the whole, consistently better than students not involved in the DI approaches.
  • With the exception of affective measures, outcome estimates remained statistically and substantively significant even when a range of control variables were introduced. This means that after potential threats to the accurate and consistent interpretation of the findings were controlled statistically, the findings were still moderate to large in favor of DI approaches.
  • Strong results were consistent across all variables introduced, which is an important finding for teachers of students with learning disabilities, as it suggests DI can be effective across academic areas, grade levels, and a range of student characteristics.
  • Findings were stronger at the kindergarten level and when students participated in the DI programs for longer periods.


DI appears to be a valuable approach to teaching many students, including those with learning disabilities. DI in all academic areas is associated with moderate to large effects, and thus, teachers may have some confidence that these approaches can be effective with their students. Of course, we suggest that teachers monitor the progress of their students to assure that they are making the expected gains and adjust instruction as needed to meet the needs of their students.