Direct Instruction is an approach to teaching. It is skills-oriented, and the teaching practices it implies are teacher-directed. It emphasizes the use of small-group, face-to-face instruction by teachers and aides using carefully articulated lessons in which cognitive skills are broken down into small units, sequenced deliberately, and taught explicitly (see Carnine, 2000, pp. 5-6; Traub, 1999).

Direct instruction derives mainly from two lines of scholarship and curriculum development. One line of scholarship is based on a synthesis of findings from experimental studies (conducted by many different researchers, working independently, mostly in the 1980s) in which teachers were trained to use particular instructional practices. These practices then were assessed for their effects on student learning, and the effects were compared with effects for similar students who had not been taught according to the experimental method. The synthesis growing out of these studies identified common "teaching functions" abstracted from the experiments that had proved effective in improving student learning. These teaching functions included teaching in small steps with student practice after each step, guiding students during initial practice, and ensuring that all students experienced a high level of successful practice. Instruction of this sort was described variously by the people who used it and discussed it. It was sometimes called systematic teaching, or explicit teaching, or active teaching. In an influential essay, Barak Rosenshine and Robert Stevens (1986) called it direct instruction, and this is the name by which it is now most often known.

As Rosenshine and Stevens describe it, direct instruction is a teaching model, not a particular, fully elaborated program for teaching, say, reading or mathematics. It is abstracted from detailed procedures found, for example, in particular training manuals and materials, and it implies nothing definite about how teaching functions it embodies (Rosenshine & Stevens, 1986, p. 389). It is a generic teaching model, in other words - one awaiting subsequent interpretation and development in particular applications.

Interpretation and development of this sort has been provided in a second line of scholarship associated primarily with the work of Siegfried Engelmann and his colleagues. Their work goes beyond the generic direct instruction model, providing detailed teaching programs consistent with its main principles. Engelmann and his colleagues call their programs Direct Instruction or DI programs, using upper-case type type to distinguish from the earlier, generic formulations.

The texture of detail in Direct Instruction derives in part from its foundation in close analyses of the comprehension and reasoning skills needed for successful performance in, say, reading or mathematics. These skills provide the intellectual substance of Direct Instruction programs. In the case of reading, it is substance found in the sound system of spoken English and the ways in which English sounds are represented in writing. That is why Direct Instruction is associated with phonemic awareness, or phonics. But Direct Instruction is not the same thing as phonics, or "merely phonics." Direct Instruction can be used to teach other things other than phonics - mathematics and logic, for example - and phonics can be taught (as it often has been) by means other than Direct Instruction.

The detailed chapter of Direct Instruction derives also from a learning theory (Engelmann & Carnine, 1991) and a set of teaching practices linked to that theory. The learning theory focuses on how children generalize from present understanding to understanding of new, untaught examples. This theory informs the sequencing of classroom tasks for children and the means by which the teachers lead children through those tasks. The means include a complex system of scripted remarks, questions, and signals, to which children provide individual and choral responses in extended, interactive sessions. Children in Direct Instruction classrooms also do written work in workbook or activity sheets.

Many published instructional programs have made some use of insights from Direct Instruction (or direct instruction). Taken at a high level of generality, at least, those insights are not private property. But Direct Instruction to date is represented most clearly and extensively in instructional programs authored by Engelmann and published by SRA/McGraw-Hill.

When educators talk about adopting Direct Instruction, the programs in question are most likely the Engelmann-authored SRA/McGraw-Hill programs. Other publishers, of course, could enter the market, if they chose to do so, by developing direct instruction principles.