Published: March 17, 1999 Ed Week
A Direct Challenge
|Many experts don't like the rigid structure of the Direct Instruction reform model, but a sizable body of research says it can raise achievement.|
When an independent research group evaluated the research backing up 24 popular school reform models this year, it found two surprises.
The first surprise was that only three programs could point to strong evidence that they were effective in improving student achievement. The second surprise was that Direct Instruction, a program long scorned by many educators and academics for its lock-step structure, was one of them.
Direct Instruction grew out of studies on the teaching of beginning reading that Siegfried Engelmann began at the University of Illinois in the 1960s. Thirty years later, only 150 schools across the country use on a schoolwide basis the program he developed. By comparison, Success for All, another reform model with high marks for its solid research base, is used in more than 1,100 schools.
Thousands more schools, however, use Direct Instruction's commercially produced materials--usually in remedial classrooms, special education resource rooms, or special programs for disadvantaged students.
"We were sort of like the plague for regular education," says Mr. Engelmann, now 67 and a professor at the University of Oregon. "Regular education would have nothing to do with us. It wasn't until the last few years that we started to break the mold."
SOURCE: "Catalog of School Reform Models, First Edition," Northwest Regional Education Laboratory, 1998.
Part of what has made his program so disliked in many quarters is that it requires teachers to adhere strictly to scripted, carefully sequenced lessons. Critics often accuse the program's developers of peddling "teacher proof" curricula.
But Mr. Engelmann's theory is that students learn more if instructional presentations are clear. Those instructions, he believes, should rule out any misinterpretations and help students generalize skills in different contexts.
And Mr. Engelmann and his colleagues have devoted years of study to pinpointing where students make mistakes and how to avoid them. Some 30 experiments alone, Mr. Engelmann estimates, have focused on technical details of the program, such as the pacing of lessons.
The program also emphasizes basic academic skills. The idea is that students must master the basics before they can move on to more-complex thinking activities.
For More Information:
With Gary L. Adams, Siegfried Engelmann also conducted his own meta-analysis of 37 Direct Instruction studies. Information on ordering that report, "Research on Direct Instruction: 25 Years Beyond DISTAR," is available from the Association for Direct Instruction, PO Box 10252, Eugene, OR 97440; (541) 485-1163.
Direct Instruction's first incarnation was a reading and mathematics program for pupils in kindergarten through 3rd grade, known as DISTAR or Direct Instructional System for Teaching and Remediation. The early programs have since been expanded to reach children in prekindergarten through 6th grade. And they have grown beyond reading and math to include lessons in social studies, science, writing, reasoning, and spelling.
Dozens of studies have found over the years that in head-to-head comparisons with traditional classroom instruction or other educational interventions, the winner is often Direct Instruction or DISTAR. The largest of those evaluations was a $59 million study that from 1968 to 1976 compared 20 different programs used in the federal government's Follow Through initiative, a massive educational effort launched as part of President Lyndon B. Johnson's "war on poverty."
The researchers concluded that, of all the programs studied, Direct Instruction produced the biggest gains in students' basic skills and thinking abilities---even in their self-esteem.
Other, smaller studies also suggest that the program improves students' chances of graduating from high school and attending college.
That kind of track record helped earn Direct Instruction its high rating from the American Institutes for Research, the Washington-based group that reviewed data on two dozen "whole school" reform models in a study released last month. Of the 14 Direct Instruction studies that met AIR's standards for scientific rigor, seven found gains in reading, 11 in mathematics, and nine in language. ("Researchers Rate Whole-School Reform Models," Feb. 17, 1999.)
The American Federation of Teachers, after a similar review last year, chose Direct Instruction as one of six schoolwide programs that showed promise in raising achievement. And the program also made its way onto a list suggesting research-backed models that schools could adopt to qualify for a share of $150 million in new federal grants.
The sudden attention, though, doesn't mean that Direct Instruction has completely shed its controversial reputation.
"What happened with Project Follow Through was that it was a big, messy study," says Lawrence J. Schweinhart, the research-division chairman for the High/Scope Educational Foundation, which developed one of the preschool models compared with Direct Instruction in that study.
Mr. Schweinhart and his colleague David P. Weikart conducted their own study comparing Direct Instruction with other preschool programs. They tracked groups of poor children who had been randomly assigned to one of three different types of preschool classrooms: a Direct Instruction program, a traditional nursery school program, or their own High/Scope model, which calls for allowing children to plan and carry out their own learning activities.
All three models produced academic gains in the preschoolers, but the Direct Instruction pupils scored the highest. But in other areas of the children's development, the findings were less promising.
By age 15, 46 percent of the Direct Instruction students had been identified as having emotional problems--a significantly higher percentage than children who had been in either of the other two programs. By age 23, the former Direct Instruction preschoolers had accumulated more felony arrests.
Mr. Schweinhart says the program's authoritarian structure and lack of attention to students' social and emotional needs may help explain the results. "I don't think there is any question that Direct Instruction is a great way to improve school achievement if that were the only goal in the world," he says. "But it isn't our only goal."
Mr. Engelmann disputes those findings. He notes that, with only 68 students to begin with, the study was too small to produce reliable results. He and his colleagues have also cited other technical problems with the data.
Other critics, however, point out that many of the positive studies on Direct Instruction were conducted by researchers associated with the program--a common caveat with many education studies. But Robert Slavin, a Johns Hopkins University researcher who also favorably reviewed the data on Direct Instruction for Show Me the Evidence: Proven and Promising Programs for America's Schools, a book he wrote last year with Olatokunbo S. Fashola, says that some of the criticism about evaluator bias could be a red herring.
"If you have 20 studies--and I'm just making that number up--and four are independent, that's a small proportion. But if all four are positive, that's impressive," he says. The same criticism, in fact, has also dogged Success for All, the program that Mr. Slavin developed.
But critics of Direct Instruction also note that many of the studies that shed a favorable light on the program are more than a decade old. In today's classrooms, the same conclusions might not hold up.
"One of the problems is that to have proven programs, you have to have old programs," adds Richard L. Allington, the chairman of the reading department at the State University of New York at Albany. "Most of these Direct Instruction programs have been around 25 or 26 years, which is why there's more 'research' on them."
If Direct Instruction looks good, Mr. Allington and others say, it may be because there is a dearth of effectiveness data on anything else.
Like most recent graduates, Matthew Carpenter never heard of Direct Instruction when he was studying to be a teacher in Elmira College in Elmira, N.Y. His professors did, however, warn him against programs that emphasize the phonics method of reading instruction as strictly as Direct Instruction does.
But as a second-year teacher, Mr. Carpenter says the program's tight structure has benefited both himself and the disadvantaged students he teaches at Arundel Elementary School in Baltimore. "I like the structure," he says. "I think it's good for this group of kids."
Mr. Carpenter's 6th graders last month were working on their reasoning and writing skills. Their task: Take two sentences and make a new sentence from them that begins with the word 'no' and uses the word 'only.'
"The wolves howled and ate at night," Mr. Carpenter reads. "The wolves did not eat."
Fourteen youngsters bend over their papers, writing their answers as their teacher walks around checking their work. Direct Instruction theorists believe it's important to catch errors quickly before the mistakes imprint themselves on impressionable brains.
"The answer is...?" Mr. Carpenter prompts. The students shout out in unison. "No, the wolves only howled at night." Similar chanting is also audible from classrooms down the hall.
A Return to the Basics
After five more such exercises, the class moves on to the next task in its language arts textbook, which is to identify the parts of speech in a series of sentences. Mr. Carpenter reads the first one: "That last statement is very misleading."
"What's the noun in the subject?" he asks, reading from his script. He snaps his fingers and the students shout: "Statement!"
"What's the verb, everyone?" Fingers snap, students shout. "Is!"
"Good job," the teacher replies.
So the lessons go. Except for one hour-long period, Mr. Carpenter uses Direct Instruction all day long.
The least structured of those classes is U.S. history, when students get a chance to ask as well as answer questions and to write the kinds of longer reports and essays not included in the fast-moving writing classes. In the history class, they are asked, for example, to draw connections between the Magna Carta and protests over the stamp tax in Colonial America, and they ponder what life in the United States would be like if President Clinton were king. And this teacher and his students say it's their favorite subject.
Arundel Elementary sits atop a hill studded with housing projects and apartment complexes; 96 percent of its students come from families poor enough to qualify for federally subsidized lunches. It is one of 18 schools in Baltimore that use Direct Instruction with support from the Baltimore Curriculum Project, a private, nonprofit group. The group stumbled upon Direct Instruction in its search for a basic-skills curriculum similar to the program used by the Calvert School, a private, Baltimore-area school that has won acclaim for its high student achievement.
Now in its third year of implementing Direct Instruction, Arundel is a relative neophyte at it. And the 410-student school's scores on Maryland's tough, criterion-referenced state exams have yet to show across-the-board increases as a result of the program. Scores in the relatively young testing program have gone up and down in several poor schools, notes Muriel Berkeley, the director of the Baltimore Curriculum Project.
But Direct Instruction is credited with engineering a much-publicized turnaround at nearby City Springs Elementary School. Once considered one of the worst schools in Baltimore, City Springs now has orderly classrooms and higher test scores.
"I think the better order comes because kids are more engaged in what they are doing," Ms. Berkeley says.
Schools in Texas and Utah also credit the program with raising test scores. In what is probably Direct Instruction's biggest success story, students at one of those schools--predominantly poor Wesley Elementary School in Houston--rank in the top tier of all elementary schools in the state.
But, while nearly all of the Baltimore schools using the program serve poor students, Direct Instruction can also work in middle-class schools and in high-achieving classrooms, its proponents contend.
"We make a lot of assumptions in education," Ms. Berkeley says. "A child looks bright or a child knows how to read, but we don't consider that the child may have some missing skills."
Direct Instruction, she argues, can fill any gaps: "What this gives you is a vertebrae--a backbone--to make sure you haven't skipped any skills."
All the students in Mr. Carpenter's class, for example, are considered advanced. To move children through their lessons quickly and efficiently, Direct Instruction calls for placing students in groups based on their abilities. But the groups are also flexible, and when the class turns to U.S. history, six students go across the hall for extra language arts lessons. To enable that kind of cross-classroom grouping, schools that sign on to the program must require all their classrooms to teach the same subjects at the same time each day.
But in math, the advanced students also work ahead, outpacing Mr. Carpenter's ability to read the lessons, something the teacher says can be frustrating.
"We already know this stuff," says Troy Mouzon, a 5th grader whose exceptional performance has put him with this 6th grade group.
Such occurrences, Mr. Engelmann says, are a "no-no." Because the skills taught build on one another over time, even the most agile students need to stick with the lessons, he says.
Repetition is also part of the program's game plan. Students do not really master a skill until they repeat it again and again and in different contexts, program providers say.
Even with a script, Direct Instruction looks different in different classes. Some teachers, for example, moderate their delivery to accentuate key ideas. In other classes, the pace may be more staccato or more like a game.
Conducting the lessons properly, proponents say, can be tiring. The teachers must be "on" all day long and attend to individual students' progress. "It's like actors in a play," Ms. Berkeley says. "We don't ask the actor to write the play, but he interprets the play and presents the play."
Mr. Engelmann estimates that teachers need at least two years to teach the program well. "There are no dyslexic kids--only dysteachic teachers," he says with characteristic bluntness.
Training begins with a weeklong in-service session. In addition, Direct Instruction providers visit classrooms on a monthly basis to coach and observe.
But, unlike other kinds of professional developers, Direct Instruction coaches do not hang back in the classroom and take notes. They jump in when they see a potential problem--a tactic that rubs some teachers the wrong way. "It's like they undermine you in front of your students," says James Sarath, another 6th grade teacher at Arundel Elementary School.
Mr. Carpenter says his students once offered to beat up a Direct Instruction trainer for "disrespecting" their teacher.
Because the lesson scripts are also sold commercially by Science Research Associates, a division of Macmillan/McGraw-Hill School Publishing Co., some educators also get the idea they need no training. And that attitude has led to uneven implementation of the program nationwide.
Mr. Engelmann hopes that several studies and teacher surveys now in the works will help answer concerns about the program. But even with its newfound, higher profile, Direct Instruction still faces an uphill battle in cracking the education establishment.
When Arundel Elementary moved to adopt the program, several teachers transferred to other schools. The Denver-based Education Commission of the States, which also reviewed the program for its "Promising Practices" database, concludes that, despite an impressive research record, "Direct Instruction is not for all children under all circumstances or for all teachers or schools."
"Research or no research," adds Mr. Slavin, the developer of Success for All, "many schools would say that's just not a program that fits with their philosophy."