Direct Instruction: The Power to Change it All
Substantial proportions of our nation’s students do not meet even basic levels of academic proficiency. This has lifelong implications for students and for society as a whole. But this dire situation could change. I base this conclusion on a systematic, quantitative review of the studies of Direct Instruction conducted with colleagues Tim Wood, Cristy Coughlin, and Caitlin Khoury.
Direct Instruction (DI) is a highly structured, research-based method of instruction developed by Siegfried Engelmann and his colleagues beginning in the 1960s. It assumes all students can succeed with appropriate instruction. If students are not learning, the problem does not lie with the student but with the teaching. Based on this assumption, Englemann and his colleagues, based on years of research and testing, developed curricula designed to be effective and efficient.
Because DI was first developed in the 1960s there are many studies of its effectiveness, and my colleagues and I examined that literature using the statistical technique of meta-analysis. With help from our marvelous university librarians, we found 377 studies of DI. From these we calculated more than 4,600 estimates of DI’s effectiveness, called “effect sizes.” Because effect sizes are commonly used in research, we could compare the data from the DI literature to effects found for other curricula.
Our results were striking—so striking and so strong that we checked and rechecked the results to make sure they were correct. We also used a range of statistical methods to look for any conditions that might alter our findings. But, no matter how we sliced the data we got the same results. The effect sizes associated with DI were about twice as strong as the average found with other curricula. These strong effects appeared across more than 50 years of research, in studies with students of all types of backgrounds, in different types of schools, in studies with various methodologies, and in comparisons with a wide range of alternative curricula. The strong effects appeared with both academic and affective outcomes. Students taught with DI not only had higher academic achievement; they also had greater confidence in their ability to learn and succeed. There were only two factors that affected the magnitude of DI’s impact: dosage and fidelity. Students did even better when their teachers adhered more closely to the programs’ guidelines and when they were taught with DI for a longer period.
From these data, it seems clear the widespread use of DI could do much to increase student achievement. Yet, it is not widely used and, in fact, many in the educational community appear to be quite antagonistic toward its use.
My co-authors and I are social scientists, with backgrounds in sociology, psychology, and history. We believe empirical evidence and scientific research provide vital information for addressing difficult social problems. Thus, it is hard for us to understand the opposition to DI and the “curriculum wars.” We know these views are deeply embedded in the culture of the education profession. But we also know teachers and administrators care deeply about their students and want them to do well. We also know cultures and social institutions can change.
We sincerely hope the results of our study can help turn the tide of educational thought and policy. The data show all students can learn if they are given appropriate instruction. It is time for educators to work with policymakers and the public for change. Our students deserve no less.
Please join me Wednesday, February 21, as I present an EDVIEW360 webinar called, “Success for Every Student: The Research Behind Direct Instruction.” I’ll expand on this topic and present useful strategies, as well.