Education That Works:
The Child is Always Right, Part Five
Direct Instruction

Edward L. Anderson, Ph.D.

Ballantyne School, El Cajon CA: Third Grade

Figure 1

Figure 1 shows changes in third grade scores at the Ballantyne School, El Cajon, CA after adopting Direct Instruction in 1984. Reading percentiles improved from 32nd to 58th, math from 42nd to 70th and language from 35th to 70th. These are major changes!

DI was tested in “Project Follow Through,” a U.S. Office of Education funded Level 3 experiment to compare the effectiveness of nine different teaching models in Head Start education. It ran nine years, from 1968 through 1977, involved over 75 schools and followed thousands of students in successive waves from prekindergarten to fifth grade.⁽⁶⁾ Said to be “the largest social experiment ever performed,” it cost close to $1 billion.⁽⁷⁾ Schools in each program model were matched for nearly identical pupil characteristics with "comparison groups" which continued whatever teaching methods were already in use. These were the groups against which each model’s classes would be compared.

Figure 2, “Follow Through” Index of Significant Results, shows a vertical, zero line. It represents the average of the “comparison groups.” The further to the right the bar goes from the centerline, the higher the proportion of sites in which that model showed both educationally and statistically significant effects. Bars going to the left show negative results. Direct Instruction clearly did best and some models did even poorer than the comparison group.⁽⁸⁾

“Follow Through” Index of Significant Results

Basics = word knowledge, spelling. language. Math computation
Cognitive = reading comprehension, math concepts, math problems

Figure 2

We have discussed Direct Instruction. Parent Education was an extension of a home-based, highly individualized program. Behavior Analysis used reinforcement and a structured curriculum but not the complete DI program. Southwest Lab used programmed curricula for bilingual children, emphasizing language development. Bank Street aimed the development of the whole child, targeting self-image, creativity, coping skills and use of language. Responsive Education aimed at development of problem-solving skills, sensory discrimination and self-confidence. TEEM targeted broad intellectual skills and positive attitudes toward school. Cognitive Curriculum was based on Piaget’s theories. Children scheduled their own activities and teachers acted as catalysts rather than instructors. Open Education used the famous “open classroom” in development of self-respect, imagination and openness to change.

The bars show results for Basics, that is, reading words, word knowledge, spelling, language, and math computation; Cognitive, which is reading comprehension, math concepts and math problems; and Self-Esteem.

Figure 3 “Follow Through” Evaluation, shows the percentiles attained by the end of the third grade in four areas.⁽⁹⁾ The centerline represents the 20th percentile, the expected performance of the disadvantaged children. Note again that no model performed as well as Direct Instruction. Cognitive Curriculum and Open Education, the child-centered philosophies most favored today, performed poorer than the already disastrous 20th percentile!

“Follow Through” Evaluation

Figure 3

Competing against these eight other instructional models, DI scored first in the teaching of reading, arithmetic, spelling, language, basic skills, academic cognitive skills and, unexpectedly, positive self-image. This latter even against models which had improvement of the student’s self-image as key to the learning process.

It ranked first in urban sites, and with kids for whom English was a second language. It was especially effective with average background kids who had started the program in kindergarten. They were reading at third grade level at the end of first grade. Children with IQs from 70 to above 130 learned at approximately the same rate. The lower the IQ, the lower the base, and the lower the level achieved over the four years of instruction, but they gain nearly as much. Their IQs also improved. Might more time with the lower achievers narrow the IQ and background gap?

Despite these results, in an action unexplained to this day, all models were certified as equally effective by the Office of Education.

Reanalysis^(10,11) of the basic data has confirmed the original conclusions regarding Direct Instruction, but since then these older results have been superseded by 15 more years of DI research and demonstration. By 1981, the last of twelve DI Follow Through projects was demonstrated as effective. Eleven had 8 to 10 years of data on successive groups of children- replication over time. Schools were from large cities (New York, San Diego, St. Louis); middle sized cities (Flint, MI; Dayton, OH; E. St. Louis, IL); rural white communities (Flippin, AR; Smithville, TN); a rural black community (Williamsburg, SC); Hispanic communities (Uvalde, TX; E. Las Vegas, NM); and a Native American community (Cherokee, NC). All projects were certified as exemplary in reading and mathematics for the primary grades.

Among others, the East Las Vegas, NM, program has continued. Though it’s students are of Hispanic background with English as a second language, the school is at the top of New Mexico’s schools on standardized tests given, of course, in English. Unfortunately, over 100 other DI programs have been dropped: “this isn’t the way we do things!” Nevertheless, additional thousands of students have been taught using DI and hundreds are in DI programs today. They show the same teaching effectiveness and, as you will see, modifications to the Direct Instruction base may be doing even better.

Contrary to some current opinion, at least some benefits of Head Start persisted. Students with three or four years of DI maintained their advantage in fifth and sixth grades.⁽¹²⁾ Similarly, DI students maintained an advantage in junior and senior high school as evidenced by achievement tests, graduation rates, and college applications and acceptances.⁽¹³⁾

The use of DI by the Wesley Elementary School in Houston, TX, was featured on “Prime Time Live.”⁽¹⁴⁾ You see educationally disadvantaged ghetto kids reading and discussing Macbeth, in the original, in fifth grade. Unfortunately, most of the show was devoted to the continued opposition to the program by school administrators; for example, private funds had to be found to buy maps because third graders “weren’t ready for maps.”

Beyond elementary skills, DI has been extended to geometry, literary analysis, chemistry, critical thinking, social studies, US history and music, though with many fewer demonstrations. DI has been highly successful with the developmentally disabled, with hundreds of replications and thousands of students. Unpublished, preliminary data show that the methods are also very effective with the educationally gifted.

Edward L. Anderson, Ph.D.
Ed Anderson was among the founders of the Cambridge Center,
and remains one of its most active supporters.
This paper is based upon a talk that Dr. Anderson gave on August 2, 1994,
sponsored by the Chautauqua Institution, Chautauqua, New York.

Return Home