THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 DIRECT INSTRUCTION: AN EDUCATORS’ GUIDE AND A PLEA FOR ACTION THOMAS KIM AND SAUL AXELROD TEMPLE UNIVERSITY The achievement gap between minority youth and the majority population in the United States is well documented (Johnston & Viadero, 2000). In order to reduce this gap and avail children of poverty of the most effective educational techniques, Project Follow Through was conducted in the 1960’s. The research that compared a variety of educational approaches found Direct Instruction techniques to be the most effective along all measures. Since that time there have been numerous studies that have demonstrated the great benefit of Direct Instruction procedures for the education of minority youth. Yet, few school systems use the procedures. The article discusses arguments for and against the adoption of Direct Instruction techniques and makes a plea for minority educators to lead a movement that will lead to widespread espousal of the procedures. Key words: direct instruction, academic achievement, dissemination, inner - city students. THE SCOPE OF THE PROBLEM suburbs the Black-White achievement gap has been docu- Since the 1970’s the National Assessment of Educa- mented to persist (Johnston & Viadero, 2000). tional Progress (NAEP)—also popularly known as the At the same time, Black and Latino students are mak- Nation’s Report Card—has been assessing K-12 students ing up a larger and larger percentage of the school-age throughout the U.S. in their skills across common aca- population. According to the U.S. Bureau of the Census, demic disciplines. These scores are then differentiated Black and Latino children will make up 34% of the by subgroups such as age, ethnicity, socioeconomic sta- school-age population in 2010. In many urban school dis- tus, and gender to show comparative trends in academic tricts, these groups already comprise more than 80% of performance. From the time these and other similar as- the student population (Kober, 2001). Furthermore, el- sessments have been enacted, a significant disparity has ementary and high school academic achievement scores been apparent between the academic achievement of have been shown to correlate strongly with high school White students and students in minority groups—Black, completion, college enrollment, and labor market out- Latino, and Native American, specifically. This “achieve- comes (Jencks & Phillips, 1998; Johnston & Viadero, ment gap” has been well-documented and tracked care- 2000; NCES, 2001). Recent studies have found that the fully over the years; the 1970’s and 80’s showed marked academic achievement gap between Blacks and Whites progress in narrowing the gap in several important sub- could statistically account for most of the eventual wage ject areas, but this relative progress has since stalled differential between the two groups (NCES, 2001). (Johnston & Viadero, 2000; National Center for Educa- Johnston & Viadero cite data compiled by the RAND tion Statistics, NCES, 2001). Corp. that found Latino children will be 2.6 times more Kober’s (2001) analysis of the achievement gap points likely to grow up in poverty than White children in the out that the 1999 NAEP statistics in both trend assess- year 2015, a ratio that has steadily increased over the ments and main assessments reveal a persistent disparity past decade. These repercussions are compounded by the between the academic achievement of White majority stu- changing marketplace, which demands more educated dents and Black, Latino, and Native American minority workers for a society revolutionized by technological students. At every age and subject level, Black and Latino changes and worldwide globalization. When America’s students trailed White students—sometimes by several shifting demographic is considered, along with the im- equivalent academic years or grade levels. In the 1998 plications of early academic achievement, the achieve- NAEP main writing assessment, 8% of Black 4th graders ment gap phenomenon has consequences beyond the dis- and 10% of Latino 4th graders scored at a Proficient level, enfranchisement of a few minority groups to the fate of a compared to the 27% of White 4th graders. Similarly, the national economy. NCES reported in 2001 that Black students overall had In their overview of the possible explanations for the lower math and reading scores than White students at achievement gap, Jencks and Phillips (1998) posit that every grade level. Even within integrated, middle-class no single factor fully accounts for the phenomenon, al- though aspects of family and cultural background do play 111 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 important roles, particularly when interacting with insti- math skills, higher order skills in cognitive and concep- tutions like schools. Indeed, several researchers have ar- tual thinking—even self-esteem (Adams & Engelmann, gued that the very existence of the achievement gap con- 1996; AFT, 2002; Becker & Carnine, 1981; Coombs, firms a systemic racism in the American educational sys- 1998; Parsons & Polson, 2000; Schuman, 2002). In other tem where Blacks and other minority groups are at a con- words, Direct Instruction, a model classified as teacher- sistent institutional disadvantage in the quality of their centered and basic skills-oriented, outperformed other education (Johnston & Viadero, 2000; Skrla, Scheurich, models—models deemed to be cognitive or affective in Johnson, & Koschoreck, 2001). Ferguson (1998), in his nature—not only in basic skills achievement, but in cog- survey of initiatives addressing the achievement gap, nitive and affective achievement as well. The students in found that schools themselves can help reduce the the Direct Instruction programs improved, on average, achievement gap by enacting research-proven strategies from about the 20th percentile to the 41st percentile in read- to raise academic performance. ing scores alone (Schuman, 2002; AFT, 2002). WHAT WORKS WITH CHILDREN OF POV- DOES EFFECTIVENESS MATTER? ERTY? Direct Instruction’s empirical superiority did not re- In 1967 a national study to find such strategies was sult in an enthusiastic endorsement, however. In fact, be- helmed by the Lyndon B. Johnson administration as part fore the findings were officially published, the Ford Foun- of its War on Poverty. Sponsored by the Department of dation commissioned a critique of them and the Depart- Education and conducted by the Stanford Research Insti- ment of Education eventually gave a blanket recommen- tute, Project Follow Through has been cited as the larg- dation to all of the models and programs in the study, est controlled comparative study of teaching methods regardless of academic efficacy (Watkins, 1995). Gene ever, involving about 700,000 students in 170 communi- Glass, in a critique of Project Follow Through that was ties across the United States (Bock, Stebbins & Proper, published by the National Institute of Education, claimed 1977; Nadler, 1998; Parsons & Polson, 2000; Schuman, that teachers did not “need statistical findings of experi- 2002; Watkins, 1997). It used a planned variation ex- ments to decide how best to teach children” (Glass, 1993). perimental design; parents in selected communities re- A comparable national research initiative of pedagogical viewed the proposals of over 22 educational models and practices has not been undertaken since. requested one to be implemented in an area school. Each school with an experimental implementation was matched OBJECTIONS TO DIRECT INSTRUCTION with a “control” school within the same community that TEACHING PROCEDURES would not receive any such implementation (Schuman, Why the backlash? Why were the findings of Project 2002). The study was, therefore, a general comparison of Follow Through so immediately and summarily dis- composite educational models to identify “best practices” missed? Advocates of Direct Instruction have maintained for school reform (Parsons & Polson, 2000). that the educational establishment had—and has—a Ultimately, 12 models of instruction were compared, strong philosophical bias favoring child-centered peda- including 4 major representatives of child-centered peda- gogy that is almost dogmatically held regardless of re- gogy: Constructivism or Discovery Learning, Whole Lan- search results (Carnine, 2000; Silverman, 2004). In 2002, guage, Developmentally Appropriate Practices, and the a series of focus groups and national surveys commis- Open Education Model (Bock, et al., 1977; Watkins, sioned by the Manhattan Institute for Policy Research 1997). The most requested model in the study, however, confirmed the influence of this bias among classroom was Direct Instruction, a teacher-centered approach for teachers: 56% of teachers surveyed admitted having a training academic skills (Schuman, 2002); it was imple- teaching philosophy that was more “student-directed” mented in 18 school districts for Project Follow Through than “teacher-directed” and only 15% believed it was (Nadler, 1998). important to teach students “specific information and A 1977 evaluation of the Project Follow Through re- skills.” sults revealed that scores on the Metropolitan Achieve- A child-centric approach to education demands that ment Test, Coopersmith Self-Esteem Inventory, and In- the impetus, direction, and style of instruction be guided tellectual Achievement Responsibility Scale overwhelm- by each student’s intrinsic motivation to learn and dis- ingly favored Direct Instruction in superior student cover. It is a philosophy that largely developed in reac- achievement over other models and control schools in tion to a teacher-centric or content-centric approach to nearly every comparative category: basic reading and education, where the emphasis is on the effective trans- 112 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 fer of a systematic body of knowledge or skills. Propo- tion (Gersten & Domino, 1993), which is geared toward nents of this “child-centered” or “progressive” educa- the average student. Other students—those who find the tional approach often attack the use of sequenced, struc- material and/or instruction too difficult or too simplis- tured instruction as being not only insensitive to the needs tic—are largely left to adjust or adapt as best they can. and interests of the learner, but also ultimately ineffec- Child-centered approaches provide open-ended engage- tive in promoting true intellectual development. ments for students so that they can explore curricular Kozloff, LaNunziata, Cowardin, & Bessellieu (2001), material at their own interest and aptitude; nevertheless, however, point out that “the design principles underly- students are still left with most of the responsibility for ing child-centered models like ‘constructivist,’ ‘inquiry their learning, including their level of mastery. Direct curricula,’ and ‘developmentally appropriate best prac- Instruction, on the other hand, calls for the design of an tices’ are at odds with the large body of experimental educational system that adjusts the curriculum and in- research on learning”. Child-centered approaches like struction around each student’s performance so that ev- constructivism and whole language reading instruction ery student experiences a high rate of success while ad- have been criticized for being unconcerned about aca- hering to fixed standards of achievement (Gersten & demic standards (Kozloff, et al., 2001) and remiss in pro- Domino, 1993; Gleason & Hall, 1991; Engelmann & viding actual instruction to accomplish academic tasks Carnine, 1991). The onus of success, then, has been (Gersten & Domino, 1993; McCaslin, 1989; Stahl & shifted from the student to the teaching system. Miller, 1989). Students who do not or cannot self-instruct Direct Instruction, then, is not merely an inchoate edu- themselves in these pedagogies are left to trail behind in cational philosophy or ambiguous teaching approach. It accomplishment, often developing coping strategies in is more accurately a system of teaching technologies that place of actual learning (Gersten & Domino, 1993; have been developed in the pursuit of its fundamental McCaslin, 1989). Without the mastery of basic skills, pedagogical goal. In 1976, Rosenshine introduced the these students develop what Binder (1996) calls a “cu- term “direct instruction” as part of his examination of mulative dysfluency” where early deficiencies or discrep- behaviorist teaching practices. In analyzing the teaching ancies snowball into a pattern of academic failure and model developed by Siegfried Engelmann and others, he rebellion (Kozloff, et al., 2001; Montgomery & Rossi, emphasized its use of task analysis and teacher model- 1994). This is particularly relevant to minority groups. ing. Direct Instruction has often thus been misperceived Recent studies indicate that the achievement gap between as any systematic instruction with these features (Stein, White and minority children is evident as early as kin- Carnine, & Dixon, 1998). For the purposes of this ar- dergarten (West, Denton, & Reaney, 2000), and that half ticle, however, Direct Instruction refers specifically to the disparity in Black-White scores at 12th grade can be the teaching model developed by Engelmann and his col- attributed to differences already present in the first grade leagues—a system that incorporates, but is not defined (Jencks & Phillips, 1998). Furthermore, Delpit (1988) by the practices identified by Rosenshine. points out that classrooms and schools often merely rein- Unlike most child-centered models, Direct Instruction force the values and standards of the mainstream social evolved out of work with students at-risk (Becker & culture and the schools’ own institutional culture. Lack- Carnine, 1981; Bereiter & Engelmann, 1966) just as much ing explicit instruction, students in the minority or stu- pioneering work in behavioral instruction grew out of dents at-risk face steep obstacles in succeeding in such work with students with social, emotional, or mental dis- environments. Put another way, child-centered pedagogies advantages or disabilities (Gardner, et al., 1994; Ullmann are so dependent on the intuitions of the child that they & Krasner, 1966; Ulrich, Stachnick, & Mabry, 1970). Out ultimately hurt the child whose intuitions are at odds with of this work, and guided by the general philosophical ap- the norms in the school and in the society at large. proach of Direct Instruction, evolved principles of cur- ricular design, teaching strategies, classroom manage- WHAT IS DIRECT INSTRUCTION? ment, student assessment, and teacher training (Stein, et If child-centered educational systems are revolution- al., 1998). These principles, in turn, have guided the con- ary reactions to conventional basal instruction, Direct In- struction of the commercial materials that are at the heart struction is a radical reform of them. Instead of attacking of Direct Instruction implementation in schools and class- the philosophical underpinnings or implications of con- rooms (Gleason & Hall, 1991). Direct Instruction, there- ventional instruction, Direct Instruction focuses on im- fore, has several dimensions, and a full grasp of Direct proving its efficacy. Direct Instruction advocates criti- Instruction requires an understanding of the interplay cize the scattershot approach of common basal instruc- among its several components. 113 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 Direct Instruction begins with a clear and systematic low in presenting content. This scripted format is meant presentation of knowledge. To this end, the curriculum is to ensure what in Direct Instruction circles is termed subjected to a rigorous analysis to determine what is “faultless communication”: a presentation that is concise, needed to be learned and how it can be learned in a logi- consistent, unambiguous, and logical even in terms of cal and systematic manner. This analysis deliberately does the language used (Gleason & Hall, 1991; Stein, et al., not assume background knowledge, but determines how 1998). The scripted format also ensures the application to instruct prerequisite knowledge explicitly while link- of instructional strategies characteristic of the Direct In- ing it to new material (Gleason & Hall, 1991; Stein, et struction methodology: active student participation, posi- al., 1998). In other words, children who arrive in the class- tive reinforcement, brisk pacing, explicit instruction, room with a disadvantage in background knowledge, guided practice, distributed review, and constant feed- whether through cultural or personal reasons, are taken back (Gleason & Hall, 1991; Nakano & Kageyama, 1992; into account at the very onset in the development of the Stein, et al., 1998). curriculum. The typical Direct Instruction lesson involves 8 to 12 The curriculum is organized around generalizable con- students actively responding to scripted teacher instruc- cepts and skills (Kameenui & Carnine, 1998; Stein, et tion for 30 to 45 minutes. Teacher-directed prompts gen- al., 1998)—ideas and learning strategies determined to erate 3 to 20 responses a minute from every student; the have the broadest application and most fundamental im- entire group often responds in unison to specific direc- pact within and across academic disciplines. Indeed, while tions that a teacher has just previously modeled. Instruc- Direct Instruction is most often implemented within tion is brisk and intensive; though the teacher is follow- schools to address deficiencies in one or two basic skills ing a script, he or she is not disengaged from the instruc- (reading or math), its curriculum is designed to be highly tion, but is constantly monitoring the class, seeking out interdisciplinary, with different strands of knowledge and responses, giving feedback, and directing behavior. Of- skills interwoven across subject areas (Kameenui & ten this teacher-directed format is followed by indepen- Carnine, 1998; Stein, et al., 1998). dent and small group work to provide additional practice The curriculum is also designed to be presented in a and application (Kozloff, et al., 2001). specific sequence so that new knowledge is built upon Students are grouped by ability; however, the group- the review, application, and mastery of older knowledge ing is flexible and dynamic (Grossen, 1996; Lott, 1998; in a manner that is clear, explicit, and manageable Miller, 2001). Students are constantly assessed: at the (Gleason & Hall, 1991). The sequence of instruction is beginning of the program, during classroom instruction, carefully designed to hold students’ attention with new and within periodic formal assessments (Gleason & Hall, knowledge while providing extensive review in a num- 1991). The curriculum is so designed that students can ber of different forms (Gleason & Hall, 1991; Stein, et shift easily to different performance groups based on their al., 1998). The method of instruction is also sequenced success rate within their own particular group. Because to provide a gradual transition from a teacher-guided for- students are always placed at a level where they are en- mat to more independent learning (Becker & Carnine, joying around a 90% success rate (Gleason & Hall, 1991), 1981; Kameenui & Carnine, 1998; Stein, et al., 1998). performance measures are not punitive but corrective. The methods of instruction are as explicitly delineated As Kozloff, et al. writes: as the content in Direct Instruction in order to ensure a faultless, efficient, and engaging presentation that is DI confronts head-on the fact of real differences in students’ tightly linked to the material presented. Based on the re- background preparation and the right of all students to achieve. search of best practices and common deficiencies in in- It does this by providing instruction tailored to the identified strengths and needs of the students…. Therefore, all students struction (Berliner, 1985; Brophy & Good, 1986; Duffy, have a maximum chance of learning all the material. All can 1983; Gunter, Denny, Jack, Shores, & Nelson, 1993; succeed (2001, p. 69). Hunter, 1980; Nelson & Johnson, 1996; Rosenshine& Stevens, 1986; Shores, et al., 1993; Stallings, 1980), Di- Direct Instruction’s emphasis on student assessment rect Instruction provides intensive instructional training— not only ensures individually appropriate instruction, but both before and during classroom implementation—as also student mastery. Coupled with the guided practice well as specific guidelines for instruction within the cur- and review intrinsic to its sequenced curriculum, Direct riculum (Gleason & Hall, 1991; Stein, et al., 1998). Di- Instruction’s system of assessment guarantees fluency rect Instruction is popularly known as a “scripted” cur- rather than mere familiarity with the material and skills riculum where teachers are given a precise script to fol- taught (Kozloff, et al., 2001). 114 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 The entire Direct Instruction curriculum is field-tested the highest percentage of students on free and reduced before it is made commercially available and then peri- lunch, the second highest student mobility rate, and the odically field-tested thereafter (Goral, 2001). Instructional highest percentage of new or inexperienced teachers. scripts, in particular, are revised constantly to ensure that In 2002 the Journal of Education for Students Placed most students within a particular performance group At-Risk dedicated its seventh issue to a series of articles achieve a 90% success rate when instructed according to analyzing the effectiveness of recent Direct Instruction the script (“Directing Direct Instruction,” 1997). Since programs in schools in Baltimore, MD, Broward County, Project Follow Through, Direct Instruction materials have FL, Fort Worth, TX, and Houston, TX. Though several been developed for most common disciplines through- of these studies were hampered by imperfect implemen- out the K-12 grades, including reading, writing, math, tations (Johnston & Viadero, 2002;Silbert, 2002) or in- science, social studies, and higher-order thinking (Adams complete analyses (Rosenshine, 2002), all of them re- & Engelmann, 1996; Kameenui & Carnine, 1998). ported substantial gains in student achievement. Schools Direct Instruction programs also insist on extensive in the Broward County study also reported improvement training to prepare for teacher implementation. Though in student behavior. Most of the schools in the studies Direct Instruction is designed to be practical and straight- were designated for the Direct Instruction programs be- forward, teachers need to become thoroughly familiar cause they were deemed to be underperforming schools with the system in order to effectively and confidently with an at-risk student population. In several instances, a apply it in their classrooms (Lott, 1998; “Directing Di- school would show particularly dramatic levels of im- rect Instruction,” 1997). In addition to preservice train- provement; City Springs Elementary School in Baltimore, ing, Direct Instruction provides supervising coaches to MD, for example, went from having its district’s lowest periodically observe and correct classroom instruction, reading scores to its fifth highest (Chenoweth, 2003). until trainee teachers are comfortable enough with the Similar case studies have been reported in schools system to apply it independently (Chenoweth, 2003; from Pennsylvania to California (Goral, 2001; Graves, Gleason & Hall, 1991). Many teachers go on to become 2002; Miller, 2001; Wilson, 2003). In addition, a number coaches themselves, mentoring their colleagues in the of comparative studies over the years have continued to system. find Direct Instruction an effective model of instruction. Adams’ and Engelmann’s 1996 review of 34 separate stud- HOW WELL DO DIRECT INSTRUCTION PRO- ies showed that the experimental results favored Direc- CEDURES WORK? tion Instruction 87% of the time; other types of instruc- Does the Direct Instruction system work? Several fol- tion only received favorable results approximately 12% low-up studies to Project Follow Through have shown of the time. The American Federation of Teachers (AFT, that students involved in Direct Instruction programs 2002) highlighted Direct Instruction as one of six prom- during Project Follow Through have continued to out- ising school wide programs in 1998. A 1999 longitudinal perform their counterparts in the control schools (Becker review of 24 school reform models (Herman, 1999) found & Gersten, 1982; Gersten, Keating, & Becker, 1998; Direct Instruction as only one of three that consistently Meyer, 1984). Former Direct Instruction students had improved student achievement. More recently, a report higher rates of high school graduation and college ac- by the Pacific Research Institute (Izumi, Coburn, & Cox, ceptance (Darch, Gersten, & Taylor, 1987; Meyer, 2002) revealed that many of the approximately 20 Cali- Gersten, & Gutkin, 1983). Thaddeus Scott Lott, the former fornia schools ranked in the top 40% of the State’s Aca- principal of one of these Direct Instruction schools— demic Performance Index that have more than 80% of Mabel B. Wesley Elementary in Houston, TX, was in- their student population in free or reduced-price lunch vited in 1998 to give a Congressional testimony about programs use a Direct Instruction program or a program his school’s success with Direct Instruction teaching pro- with similar methodologies. cedures (Lott, 1998). He testified that by 1979 students A review of the research regarding Direct Instruction in the Direct Instruction program had grade equivalent in Stein, et al. (1989) found that findings remained con- scores 1.5 t0 2.0 years higher than students prior to the sistent regardless of setting or grade level. Direct Instruc- Direct Instruction program. In the fall of 1997, Wesley tion has been found to be effective both in general edu- Elementary was one of only 13 schools in the Houston cation (Rosenshine & Stevens, 1986) and in special edu- Independent School District whose first graders scored cation (Algozzine & Maheady, 1986). It has been shown in the top 80% of the national norm in the Stanford 9 to reduce disruptive classroom behavior (Nelson & reading test. Of those 13 schools, Wesley Elementary had Johnson, 1996) as well as increase student achievement. 115 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 ment of children (DeVries, Haney, & Zan, 1991; DeVries, A RESPONSE TO CRITICISMS OF DIRECT Reese-Learned, & Morgan, 1991). A New York Times INSTRUCTION article proclaimed that Direct Instruction was an “early Nevertheless, Direct Instruction has continued to draw education pressure cooker” that led to violent behavior the ire and criticism of educators and academics, who (Hechinger, 1986). These early reactions to the study be- have discounted the research to snipe about its core peda- lie the romanticization of childhood which is at the heart gogical design. Most of these critiques are of a qualita- of child-centered pedagogical philosophies. Children are tive nature; Direct Instruction is cast in pejorative terms seen to need a natural development free from undue in- such as, “dehumanizing,” “robotic,” and “rigid” tervention or direction; to do otherwise is tantamount to (Chenoweth, 2003; Garza, 2003). Cazden (1983) com- stunting their growth and robbing their souls. Within this plained that Direct Instruction “can only be implemented paradigm, the teacher-directed structure of the Direct In- in an authoritarian, manipulative, bureaucratic system.” struction classroom represents a depersonalized authori- A close examination of the few research studies that tarian society—something children must build natural find Direct Instruction ineffective or problematic reveals defenses against to maintain their individualism and in- much about the fears and misperceptions Direct Instruc- ternal moral compass. tion generates among this vocal majority of the educa- A more recent study by Ryder, Sekulski, & Silberg tional establishment. Among the most oft-cited studies (2004) found that early elementary students performed critiquing Direct Instruction is Schweinhart, Weikart, & less well in reading comprehension within a Direct In- Larner’s research on the effect of Direct Instruction on struction program than in an open-ended classroom cur- preschool children’s social abilities (1986b). In this study riculum. The Ryder study has been criticized for many of a sample of poor children was randomly assigned to one the same design flaws as the Schweinhart study - a lim- of three preschool programs: a traditional nursery school, ited scope that could skew results, incomplete data, a a child-centered program designed by High/Scope, and a flawed Direct Instruction implementation, and possible program roughly following Direct Instruction principles. imbalance in matching students from the experimental Although students in the Direct Instruction program out- and control groups (Manzo & Park, 2004). Nevertheless, performed the others academically, more students in the the study has also generated incendiary reactions against Direct Instruction program were shown to have emotional Direct Instruction (Manzo & Park, 2004). In his conclu- problems and more were engaged in juvenile delinquency sions, Ryder recommended that in teaching poor urban by the age of 15. Schweinhart, et al. (1986b) posited that children, classrooms should enact such child-centered the authoritarian, directive nature of Direct Instruction practices as allowing students to explore personal inter- prevented students from learning how to regulate them- ests, developing projects with student input, and person- selves and discovering how to socialize with others. alizing instruction to student experiences while avoiding A follow-up study by Mills, Cole, Jenkins, & Dale Direct Instruction practices such as monitoring student (2002) found that when addressing the common criticisms time on task. of the Schweinhart study—having too small a sample size, The underlying perception, of course, is that the not representing a genuine commercial program of Di- instructivist approach of Direct Instruction is limited in rect Instruction, not fully randomizing assignments effectiveness to reasoning and skills of the lowest order. (Bereiter, 1986; Gersten, 1986; Viadero, 1999)—they Although Direct Instruction has been shown to improve could not replicate Schweinhart’s results. Instead, their overall reading scores, Ryder and others (e.g., Heshusius, prospective longitudinal study found no significant dif- 1991; Palincsar, David, Winn, & Stevens, 1991) imply ference in eventual rates of juvenile delinquency between that Direct Instruction achieves little beyond the mere the attendees of the Direct Instruction preschool program decoding of words on the page. Direct Instruction is seen compared to those of a child-centered preschool program. to promote mechanistic tasks and applications over com- In their conclusion, Mills, et al. showed that this discrep- plex cultural activities that require reflection, experimen- ancy can easily be explained when accounting for gen- tation, and personal exploration; the individual child is der differences in the sample populations; Schweinhart’s again subverted to the lockstep demands of a program- study had a disproportionate number of boys in the Di- matic social order. A more insidious implication is that rect Instruction program than in the other programs. urban children may gain skills in Direct Instruction pro- Despite the design flaws in Schweinhart, et al.’s 1986b grams but at the cost of their ability to discern and criti- study, others readily cited its results as evidence of Di- cally think about the world around them. rect Instruction’s harmful effect on the social develop- Such fears and assumptions, however, lack evidential 116 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 proof. There have been no consistent findings that reveal bers of the school community (“Directing Direct Instruc- the depression of esteem, social development, ethical de- tion,” 1997; Goral, 2001). Teachers, in particular, often velopment, critical thinking, cognitive ability, or cultural balk at the scripted format of Direct Instruction and, with- participation through Direct Instruction. Stein, et al. out proper training, will veer from the script as they see (1998) argue that many of the assertions against Direct fit (Lussier, 2003). Furthermore, administrators may find Instruction contain a fundamental confusion between rote it difficult to receive the funding for commercial Direct instruction and explicit instruction. Scripted Direct In- Instruction materials; Direct Instruction curricula have struction lessons are not based on the mass memoriza- consistently been rejected as approvable textbooks in both tion of arbitrary facts. Instead, a fundamental design prin- Texas and California (Lott, 1998; Wilson, 2003), the two ciple within the Direct Instruction curriculum is the con- largest—and among textbook publishers, most influen- veyance of generalizable strategies and concepts, though tial—Boards of Education in the country (Ravitch, 2003). this is done in an explicit and sequenced manner with From among the successful implementations of Di- constant review and assessment to ensure mastery. rect Instruction, however, several lessons can be learned: An analogy can be drawn to the mastery of chess, a 1. Administrators must be thoroughly versed in the highly complex activity that requires both skill and intu- program and its advantages (Lott, 1998). A successful ition. Instead of learning the play of chess from count- Direct Instruction implementation needs an advocate at less games with little knowledge of even the basic rules, the highest possible level to defend its use and effective- the Direct Instruction approach proposes that the rules ness against the inevitable criticisms it will draw. Such and basic strategies be taught and modeled explicitly and an advocate needs to be aware of the data that support in a manner that builds logically from simplified scenarios the program and can drive the reform (Kozloff, et al., to more complex puzzles of stratagem. Far from being 2001). unnatural, such methodologies of mimicry, practice, and 2. Administrators need to carefully consider what ad- logical progression are timeworn practices of social in- justments need to be made to accommodate the Direct struction. The guidance of a systematic sequence does Instruction program (Lott, 1998; Stein, et al., 1998). Be- not rob autonomy or reflection, but concentrates it on the sides purchasing concerns, consideration must also be skill or concept at hand. given to such issues as performance groupings, small Similarly, the scripted format of Direct Instruction group instruction, the coordination of teacher training, does not replace the creativity, initiative, and acumen of implementation evaluation, and so on (Kozloff, et al., the teachers, but frees them from the technical complexi- 2001). ties of maintaining a consistent and logical pattern of in- 3. Teachers need adequate training (Lott, 1998; struction (Goral, 2001; Stein, et al., 1998). With an in- Viadero, 1999). University teacher preparation programs structional script that is research-based and field-tested, provide, if anything, a child-centered bias that works teachers are allowed to pay more attention to their stu- against the Direct Instruction system. Engelmann, the pri- dents and respond to them in a manner that is more timely, mary architect of the Direct Instruction model, estimates helpful, and positive. A number of anecdotal teacher tes- that it takes teachers about two years to master the Direct timonies reveal that teachers often find their classroom Instruction classroom approach (Viadero, 1999). A com- experience more rewarding and meaningful because of mon problem among Direct Instruction implementations Direct Instruction’s effectiveness with all types of stu- is the lack of appropriate training before and during the dents (“Directing Direct Instruction,” 1997; Goral, 2001; use of Direct Instruction in the classroom. Miller, 2001). 4. Reforms need to be made gradually and with the confirmation of measured success (Kozloff, et al., 2001). A PLEA FOR ACTION FROM MINORITY EDU- Implementation initiatives should include a component CATORS of record-keeping and performance measurement to track Nevertheless, according to Viadero (1999), only about the program’s success. Changes should be made gradu- 150 schools across the United States use a Direct Instruc- ally and systematically, bolstered by such evidence. tion program. Much of this resistance comes from a In his article “Why education experts resist effective misperception about the nature and effects of Direct In- practices,” Carnine (2000) cites the work of Theodore struction, either from administrators, teachers, or parents. M. Porter (1996) in examining the maturation of profes- Often a Direct Instruction program is demanded by some sions and academic disciplines: party familiar with its research-tested effectiveness with- out the proper understanding and buy-in from other mem- An immature profession is characterized by expertise based on the subjective judgments of the individual professional, trust 117 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 based on personal contact rather than quantification, and au- form that cries out to be adopted. tonomy allowed by expertise and trust, which staves off stan- dardized procedures based on research findings that use con- REFERENCES trol groups. A mature profession, by contrast, is characterized Adams, G.L. & Engelmann, S. (1996). Research on Direct Instruc- by a shift from judgments of individual experts to judgments tion: 25 years beyond DISTAR. Seattle, WA: Educational Achieve- constrained by quantified data that can be inspected by a broad ment Systems. audience, less emphasis on personal trust and more on objec- Algozzine, R., & Maheady, L. (Eds.). (1986). In search of excellence: tivity, and a greater role for standardized measures and pro- instruction that works in special education classrooms. Exceptional cedures informed by scientific investigations that use control Children, 52, 487-589. groups. (p. 9). AFT (American Federation of Teachers). Six Promising Schoolwide Programs For Raising Student Achievement: Direct Instruction. (2002). Retrieved November 24, 2003, from American Federation Carnine goes further to argue that education remains of Teachers web site archived at: http://web.archive.org/web/ an immature profession, subject to the personal prefer- 20021219002326/www.aft.org/edissues/whatworks/six/di/. ences and philosophical biases of an influential cadre of Barnes, C. (2002). What do teachers teach? A survey of America’s experts and professionals. The fate of Project Follow fourth and eighth grade teachers. Center for Civic Innovation: Man- hattan Institute. Through illustrates this, as do other instances of the re- Becker, W., & Carnine, D.W. (1981). Direct Instruction: A behavior sistance Direct Instruction has faced from the educational theory model for comprehensive educational intervention with the establishment. Because of their professional influence disadvantaged. In S.W. Bijou & R. Ruiz (Eds.), Behavior modifica- over schools, teachers, and policies, advocates of child- tion: Contributions to education (pp. 145-210). Hillsdale, NJ: centered pedagogies have been able to turn a blind eye to Lawrence Erlbaum Associates. Becker, W.C. & Gersten, R. (1982). A follow-up of Follow Through: the research showing the effectiveness of Direct Instruc- The later effects of the Direct Instruction model on children in fifth tion over their own practices. and sixth grades. American Educational Research Journal, 19, 1, The growing interest in standards-based reform, how- 75-92. ever, has drawn attention to the realities of the achieve- Bereiter, C. (1986). Does Direct Instruction cause delinquency? Early Childhood Research Quarterly, 1, 289-92. ment gap: that there is a disheartening percentage of Bereiter, C. & Engelmann, S. (1966). Teaching disadvantaged chil- underperforming students within Black and Latino dren in the preschool. Engelwood Cliffs, NJ: Prentice-Hall. ethnicities, that minorities continue to trail the White Berliner, D.C. (1985). Effective classroom teaching: The necessary mainstream in academic achievement, and that this but not sufficient condition for developing exemplary schools. In achievement gap directly translates into the perpetuation G.R. Austin & H.Garber (Eds.). Research on exemplary schools (pp. 127-154), Orlando: Academic Press. of poverty and disenfranchisement for a growing segment Binder, C. (1996). Behavioral fluency: Evolution of a new paradigm. of the American population (Kober, 2001). The mount- The Behavior Analyst, 19, 163-197. ing evidence has made this problem undeniable. Howard Bock, G., Stebbins, L. & Proper, E. (1977). Education as experimen- (2003) maintains that the solution must also be data- tation: A planned variation model (Volume IV-A & B). Effects of follow through models. Washington, DC: Abt Associates. driven; research and student assessment data must shape Brophy, J. & Good, T. (1986). Teacher behavior and student achieve- curriculum and instructional strategy. He argues that ment. In M. Wittock (Ed.), Third handbook of research on teaching though educators are often averse to the accountability (pp. 328-375). Chicago, IL: Rand McNally. and algorithmic strategy such empiricism brings, true Carnine, D. (2000). Why education experts resist effective practices professionals are not put off by the numbers. They see (and what it would take to make education more like medicine). Thomas B. Fordham Foundation. Retrieved November 24, 2003, them, instead, as a tool to target their students’ needs and from http://www.edexcellence.net. address their deficiencies. In other words, they go with Cazden, C.B. (1983). Can ethnographic research go beyond the status what works. quo? Anthropology and Education Quarterly, 14, 33-41. The fact that minority children’s’ academic achieve- Chenoweth, K. (2003, December 4). Direct Instruction gets direct re- sults. The Washington Post, p. T06. ment scores in the United States do not match those of Coombs, M.K. (1998, March 24). Honest follow-through needed on the majority of children is a cause of great concern. The this project. Washington Times. Retrieved November 24, 2003, from fact that teaching techniques exist that can reduce or elimi- http://www.mathematicallycorrect.com/honestft.htm. nate this discrepancy, but are not used, is a cause for an- Darch, C., Gersten, R., & Taylor, R. (1987). Evaluation of Williamsburg ger. The purpose of the article is to arouse the interest of County Direct Instruction program: Factors leading to success in rural elementary programs. Research in Rural Education, 4, 111- minority educators in an approach to teaching that has 118. repeatedly and enduringly benefited minority children. Delpit, L.D. (1988). The silenced dialogue: Power and pedagogy in The dream of the authors is that minority educators, educating other people’s children. Harvard Education Review, 58, whether teachers, school administrators, or professors of 280-298. DeVries, R., Haney, J.P., & Zan, B. (1991). Sociomoral atmosphere education, will provide leadership to an educational re- in Direct Instruction, eclectic, and constructivist kindergartens: A 118 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 study of teachers’ enacted interpersonal understanding. Early Child- Pacific Research Institute. Retrieved February 7, 2004 , from the hood Research Quarterly, 6, 449-471. Heartland Institute: www.heartland.org/Article.cfm?artId=10733 . DeVries, R. Reese-Learned, H., & Morgan, P. (1991). Sociomoral Jencks, C., & Phillips, M. (Eds.) (1998). The Black-White test score development in direct-instruction, eclectic, and constructivist kin- gap. Washington, DC: Brookings Institution Press. dergartens: A study of children’s enacted interpersonal understand- Johnson, J. & Immerwahr, J. (1994). First things first: What Ameri- ing. Early Childhood Research Quarterly, 6, 473-517. Directing cans expect from the public schools. New York, NY: Public Agenda. Direct Instruction. (1997, October). Education Digest, 63 (2), p. Johnston, R.C., & Viadero, D. (2000, March 15). Unmet promise: 58-62. Raising minority achievement. Education Week. Duffy, G.G. (1983). From turn-taking to sense-making: Broadening Kameenui, E.J. & Carnine, D.W. (1998). Effective teaching strategies the concept of teacher effectiveness. Journal of Educational Re- that accommodate diverse learners. Upper Saddle River, NJ: Merrill. search, 76, 134-139. Kober, N. (2001). It takes more than testing: Closing the achievement Engelmann, S. & Carnine, D. (1991). Theory of instruction: Prin- gap. Center on Education Policy. Retrieved September 15, 2004 ciples and applications (Revised edition). Eugene, OR: ADI Press. from www.ctredpol.org/improvingpublicschools/ Ferguson, R.F. (1998). Can schools narrow the black-white test score closingachievementgap.pdf gap? In C. Jencks and M. Phillips (Eds.), The black-white test score Kozloff, M.A., LaNunziata, L., Cowardin, J., & Bessellieu, F.B. gap (pp. 318-374). Washington D.C.: Brookings Institution. (2001). Direct Instruction: Its contributions to high school achieve- Gardner, R., Sainato, D.M., Cooper, J.O., Heron, T.E., Heward, W.L., ment. High School Journal, 84 (2), 54-72. Eshleman, J.W., & Grossi, T.A. (1994). Behavior analysis in edu- Lott, Sr., T. S. (1998, March 26) Transcript of Testimony Committee cation. Pacific Grove, CA: Brooks/Cole Publishing Company. on Small Business Creative Approaches to Urban Education. Capi- Garza, C.L. (2003, April 19). Teaching method facing review in Duval: tol Hill Hearing Testimony, Washington, DC. Federal Document Direct Instruction’s scripted format has advocates, foes. Florida Clearing House. Times-Union, p. B-1. Lussier, C. (2003, November 17). Study questions school programs: Gersten, R. (1986). Response to consequences of three preschool cur- EBR spent $2 million, can’t document value. The Advocate (Baton riculum models through age fifteen. Early Childhood Research Rouge, LA). Quarterly, 1, 293-302. Manzo, K.K. & Park, J. (2004, January 1). Study challenges direct Gersten, R. & Domino, J. (1993). Visions and revisions: A special reading method. Education Week, 23(20), 3. education perspective on the whole language controversy. Reme- McCaslin, M. (1989). Whole language: Theory, instruction, and fu- dial & Special Education, 14 (5), 5-14. ture implementation. Elementary School Journal, 90, 223-229. Gersten, R., Keating, T., & Becker, W.C. (1988). Continued impact of McGraw-Hill Education. Results with Corrective Reading. the Direct Instruction model: Longitudinal studies of Follow Through www.mheducation.com students. Education and Treatment of Children, 11, 318-327. Meyer, L. (1984). Long-term academic effects of the Direct Instruc- Glass, G. (1993, August-September). Research news and comment— tion Project Follow Through. Elementary School Journal, 84, 380- a conversation about educational research priorities: A message to 394. Riley. Educational Researcher, 22 (6), 17-21. Gleason, M.M. & Meyer, L, Gersten, R., & Gutkin, J. (1983). Direct Instruction: A Project Hall, T.E. (1991). Focusing on instructional design to implement a Follow Through success story in an inner-city school. Elementary performance-based teacher training program: The University of School Journal, 84, 241-252. Oregon model. Education & Treatment of Children, 14, 316-333. Miller, R.J. (2001, November 28). Controversial teaching technique Goral, T. (2001). The fight about reading. Curriculum Administrator, winning fans in Sto-Rox Schools. Pittsburgh Post-Gazette, West 37 (5), 35-40. Edition, p. W-7. Graves, B. (2002, December 18). Charter school’s contract, name gives Mills, P.E., Cole, K.N., Jenkins, J.R., & Dale, P.S. (2002). Early ex- Portland pause. The Oregonian, p. B04. posure to Direct Instruction and subsequent juvenile delinquency: a Grossen, B. (1996). How shall we group to achieve excellence with prospective examination. Exceptional Children, 69, 85-97. equity. Eugene, OR: University of Oregon. Montgomery, A.F. & Rossi, R.J. (1994). Becoming at risk of failure Gunter, P.L., Denny, R.K., Jack, S.L., Shores, R.E., & Nelson, C.M. in America’s schools. In R.J. Rossi (Ed.) Schools and students at (1993). Aversive stimuli in academic interactions between students risk. New York: Teachers College Press. with serious emotional disturbance and their teachers. Behavioral Nadler, R. (1998, June 1). Failing grade. National Review, 50 (10), Disorders, 18, 265-274. 38-40. Hechinger, F.M. (1986, April 22). Preschool programs. New York Nakano, Y. & Kageyama, M. (1992). Using Direct Instruction to im- Times, 17. prove teacher performance, academic achievement, and classroom Herman, R. (1999). An educators’ guide to schoolwide reform. Ar- behavior in a Japanese public junior high school. Education & Treat- lington, VA: Educational Research Service. http://www.aasa.org/ ment of Children, 15, 326-344. (American Institutes for Research) National Center for Education Statistics (2001). Educational achieve- Heshusius, L. (1991). Curriculum-based assessment and Direct In- ment and Black-White inequality. U.S. Department of Education. struction: Critical reflections on fundamental assumptions. Excep- Retrieved September 14, 2004 from http://nces.ed.gov/pubsearch/ tional Children, 57, 315-328. pubsinfo.asp?pubid=2001061 Howard, J. (2003). “Still at risk: the causes and costs of failure to Nelson, J.R. & Johnson, A. (1996) Effects of Direct Instruction, co- educate poor and minority children for the twenty-first century.” A operative learning, and independent learning practices on the class- Nation Reformed? American education 20 years after A Nation At room behavior of students with behavioral disorders: A compara- Risk. Gordon, D.T., ed. Cambridge: Harvard Education Press. tive analysis. Journal of Emotional & Behavioral Disorders, 4 (1), Hunter, M. (1980). Teach more faster. (Rev. Ed.). El Segundo, CA: 53-63. TIP Publications. Palincsar, A.S., David, Y.M., Winn, J.A., & Stevens, D.D. (1991). Izumi, L.T., Coburn, K.G., Cox, M. (2002, November 1). They Have Examining the contexts of strategy instruction. Remedial and Spe- Overcome: High-Poverty, High-Performing Schools In California. cial Education, 12(3), 43-53. 119 THE BEHAVIOR ANALYST TODAY VOLUME NUMBER 6, ISSUE NUMBER 2, 2005 Porter, T.M. (1996). Trust in numbers: The pursuit of objectivity in West, J., Denton, K., & Reaney, L.M. (2000). The Kindergarten Year. science and public life. Princeton, NJ: Princeton University Press. Washington DC: U.S. Department of Education. Ravitch, D. (2003). The language police: How pressure groups re- Wilson, K. (2003, March 15). Effective reading program must go: strict what students learn. New York: Knopf. school’s success story isn’t on state book list. Ventura County Star, Rosenshine, B. (2002). Helping students from low-income homes read p. A01. at grade level. Journal of Education for Students Placed At Risk, 7(2), 273-283. Author Note Rosenshine, B. (1976). Recent research on teaching behavior and stu- Tom Kim is a graduate of and Saul Axelrod is a professor of dent achievement. Journal of Teacher Education, 27, 61-64. Education at Temple University. Reprints may be obtained from Rosenshine, B. & Stevens, (1986). Teaching functions. In M.C. Saul Axelrod, Department of Curriculum, Instruction, and Technol- Wittrock (Ed.), Handbook of research on teaching (Third edition) ogy in Education, Temple University, 441 Ritter Hall, Broad and (pp. 376-391). New York: McMillan. Montgomery, Philadelphia, PA 19122. Tel.: 215 - 204 – 6060. Ryder, R., Sekulski, J.L., & Silberg, A. (2004). Results of Direct In- Email:[email protected] struction reading program evaluation longitudinal results: first through third grade 2000-03. www.uwm.edu/News/PR/04.01/DI- Final-Report-2003.doc Schuman, D. (1998). Direct Instruction: A Review of the Research. Thesis from the University of North Carolina at Wilmington, Watson School of Education. Retrieved November 25, 2003 from: http:// people.uncw.edu/kozloffm/shumanthesisdi.html. Schweinhart, L.J., Weikart, D.P., & Larner, M.B. (1986a). Child-ini- tiated activities in early childhood programs may help prevent de- linquency. Early Childhood Research Quarterly, 1(3), 303-312. Schweinhart, L.J., Weikart, D.P., & Larner, M.B. (1986b). Conse- quences of three preschool curriculum models through age fifteen. Early Childhood Research Quarterly, 1(1), 15-46. Shell, E.R. (1989) Now, which kind of preschool. Psychology Today. Shores, R.E., Jack, S.L., Gunter, P.L., Ellis, D.N., DeBriere, T., & Wehby, J. (1993). Classroom interaction of children with severe behavior disorders. Journal of Emotional & Behavioral Disorders, 1, 27-39. Silbert, J. (2002). Commentary. Journal of Education for Students Placed at Risk, 7, 265-272. Silverman, J. (2004, July 5). University of Oregon professors lead national education conversation. The Associated Press State & Lo- cal Wire. Skrla, L., Scheurich, J.J., Johnson, Jr., J.F., & Koschoreck, J.W. (2001). Accountability for equity: Can state policy leverage social justice? International Journal of Leadership in Education, 4, 237-260. Stahl, S.L. & Miller, P.D. (1989). Whole language and language ex- perience approaches for beginning reading: A quantitative research synthesis. Review of Educational Research, 59(1), 87-116. Stallings, J. (1980). Allocated academic learning time revisited, or beyond time on task. Educational Researcher, 8(11), 11-16. Stein, M., Carnine, D., & Dixon, R. (1998). Direct Instruction: Inte- grating curriculum design and effective teaching practice. Interven- tion in School and Clinic, 33 (4), 227-335. Ullmann, L.P. & Krasner, L. (1966). Case studies in behavior modifi- cation. New York: Holt, Rinehart and Winston. Ulrich, R., Stachnick, T., & Mabry, J. (1970). Control of human be- havior, Volume II: From cure to prevention. Glenview, IL: Scott, Foresman. U.S. Department of Commerce. Bureau of the Census. Population Projections Program. Viadero, D. (1999, October). Scripting success. Teacher Magazine, 11 (2), 20-23. Viadero, D. (2002, April 17). Studies cite learning gains in Direct Instruction schools. Education Week, 21 (31), 15. Watkins, C.L. (1995). Follow Through: Why didn’t we. Effective School Practices, 15 (1), 5. Watkins, C. (1997). Project Follow Through: A case study of contin- gencies influence instructional practices of the educational estab- lishment. Cambridge, MA: Cambridge Center for Behavioral Stud- ies. 120