A Return to Scientific Reading Instruction
San Diego State University
San Diego, California
The importance of science in modern society is manifested in numberless ways. We look to guidance from science as to how to best protect or restore our physical or emotional well-being. Great confidence is placed on scientific solutions to everything from infant earaches to the construction of bridges and intercontinental missiles.
Knowing of the deep respect there is for scientific care of sick youngsters and for prevention of childhood diseases, it is logical to presume that science also should be consulted for answers to questions about the ways children are educated. It therefore is startling to find that many public schools today teach students to read using unscientific procedures.
Of late, reading instruction in schools has moved so far away from how relevant experimental research findings indicate it should be conducted that a vigorous controversy over this matter has developed. On one side of the “reading wars, ” as the media has dubbed the debate, are reading instruction specialists who honor experimental findings as the most suitable source of information on how to teach children to read. Lined up against defenders of scientific evidence in this regard are members of the Whole Language movement.
They are reading teaching professionals who argue that students best learn to read in the same informal, natural manner in which they previously learned to speak, as preschoolers. It consequently is held that direct, early, and systemic (DES) teaching of a prearranged hierarchy of reading skills is unnecessary. The DES teaching of reading is impractical, advocates of Whole Language teaching contend, since each child supposedly enters school with a unique, immutable learning style. In genuine Whole Language classes, a student is empowered not only to decide how he/she will learn to read, but also to personally construct the meanings of written materials.
No reputable psychologist, linguist, cognitive scientist, nor neurologist currently endorses The Whole Language hypothesis that children’s learning to speak and to read are the same linguistic processes. In addition, the effectiveness of none of the original principles nor novel practices of Whole Language reading instruction is corroborated by relevant experimental evidence. The Whole Language movement reacts to the se facts by producing much qualitative (anecdotal, nonnumerical, subjective, loosely organized, unreplicable, nonscientific) research evidence as support for its instructional innovation. This kind of evidence consistently contradicts that generated by experimental studies of children’s reading development.
The present dispute over reading instruction grew from the realization by reading teaching authorities that findings about their educational specialty from experimental, as versus qualitative research, are irreconcilable. Accordingly, anyone convinced that children must be afforded full opportunity to learn to read has to make a forced-choice between DES reading instruction based on scientific findings, and Whole Language reading teaching founded on qualitative evidence.
There is reassurance for parents, teachers, school officials, education professors, lawmakers, business and social organizations, and the public in general, who opt in favor of DES reading instruction, that they have made the correct choice. This is found in the April 2000 report by the National Reading Panel (NRP) of its critical analysis of experimental research on reading instruction (published by the National Institute of Child Health and Human Development). From a total of 300 applicants to serve as unpaid volunteers on the NRP, 14 were selected: 1 teacher, 1 school principal, 1 certified public accountant, 2 university officials, and 9 professors of education, psychology, and medicine.
The NRP’s report on how children best learn to read is based on its reference to over 100,000 applicable experimental studies on this matter. To be able to read proficiently, children must acquire four essential knowledges and skills, the NRP concluded. These are:
- Conscious awareness of the speech sounds in spoken words. This is called phonemic awareness;
- Recognition that letters are used to represent speech sounds. This is called phonics information;
- Capacity to read a text so as “to understand what is read.” To “understand” a text, the NRP believes, is to read it with the presumption that part of its meaning “resides in the intentional problem solving, thinking process of the reader.” In this view, meanings in a given text are “constructed” by each student through “a reciprocal interchange of ideas” between him/her and the message that an author intended to communicate. This is called reading comprehension.However, to many defenders of DES teaching of reading, this sort of interchange between reader and author is nothing more than imprecise reading comprehension. It is argued that personal opinions by children of the meanings authors wish to transmit are not accurate enough to be called authentic, reliable reading comprehension.
- Ability to read silently and aloud with “speed, accuracy, and proper expression.” This is called reading fluency.The NRP concedes that there is a “close relationship between [reading] fluency and reading comprehension.” This fluency also depends upon students ‘ “well developed word recognition skills.” However, “such skills do not inevitably lead to fluency,” the panel cautions. Thus, “students who are low in fluency may have difficultly getting the meaning of what they read. Here the NRP implies that a lack of fluency causes children’s inadequate reading comprehension, rather than the other way around.However, the NRP errs in asserting that “there is common agreement” among reading instruction specialists about that conclusion. To the contrary, many of these experts maintain that reading fluency depends on how well students comprehend what they read. Lack of fluency thus is judged to be a symptom, not a cause of interior reading comprehension. It consequently is held that the most effective way to improve reading fluency is to upgrade how precisely students can understand the content that authors intend to relate.Beyond the fact that the NRP describes reading comprehension and its relationship to reading fluency in a controversial manner, its report nonetheless is a faithful reflection of how the preponderance of experimental findings indicates reading instruction should be conducted. This scientific evidence discloses that:
- Development of beginning readers’ phonemic awareness (PA) should be carried out in an “explicit” manner. This instruction thus should focus “on one or two skills” of PA at a time. Moreover, speech “sounds need to be overlearned so that children can work with them automatically” (i.e., quickly and accurately). Teaching children in “small groups” is “the best way” to develop their PA.
- The “hallmark” of successful instruction of phonics information is:
· Instruction of “a planned, sequential set of phonic elements.” These are letters, speech sounds, and generalizations about how letters represent speech sounds (the alphabetic code);
· Teaching “these elements explicitly and systematically”;
· Instruction intensive enough that children “acquire sufficient knowledge and use of the alphabetic code”; and,
· Training in phonics information that “begins in kindergarten or 1st grade”
- Children’s reading fluency is best improved by teacher “guided oral reading procedures” and “feedback.” On the other hand, experimental research “has not yet demonstrated” in “a clear and convincing manner” if students’ silent reading, “individually on their own with little or no specific feedback,” has the same effect.
- Children “who are not explicitly taught” reading comprehension procedures “are unlikely to learn, develop, or use them spontaneously.” Therefore, teachers are advised they can increase students’ reading comprehension best “by explaining fully what it is they are teaching:
· what to do, why, how, and when;
· by modeling their own thinking processes;
· by encouraging children to ask questions and discuss possible answers among themselves”;
· and by assigning reading tasks “that demand active involvement” by students.
This direct and systematic method of instruction particularly is effective when students are taught “a variety of reading comprehension strategies” (techniques for gaining an understanding of meanings that authors wish to transmit).
However, the NRP is not always so straightforward in its remarks about the effect of children’s knowledge of words upon their reading comprehension. On the one hand, the panel discloses that “the finding that vocabulary [knowledge] is strongly related to [reading] comprehension seems unchallenged.”
Later on, however, the NRP deduces that a “causal link between increasing vocabulary [knowledge] and an increase in [reading] comprehension has not been demonstrated” experimentally. Despite that disclaimer, the panel feels it necessary to devote nineteen pages of its report to a description of “vocabulary instruction,” and to citations of studies made about it.
It is apparent that the NRP’s “assessment of the scientific literature on reading and its implications for reading instruction” leaves some mixed messages about teaching reading comprehension and fluency. Nevertheless, its overall impact is clear. The document clearly signals the necessity of a return to DES reading teaching that is based on experimental research findings, rather than on ideological ruminations about this instruction. Thus, in many ways the NRP report constitutes a direct confrontation to the now popular Whole Language approach to this teaching.
It is highly unlikely, however, that advocates of the Whole Language approach will react passively to the NRP’s challenge to their present eminence in the field of reading teaching. Educational organization that recommend Whole Language teaching, such as the International Reading Association and the National Council of Teachers of English, doubtless will remonstrate that the NRP report is not believable since leading members of the Whole Language movement were not selected to participate in its proceedings. The fact the panel exclusively examined scientific findings, and not qualitative evidence, also probably will be viewed with alarm.
A preview of the mode of attack upon it that the NRP report subsequently will face is found in the “Minority View” of it written by a single panel member, Joanne Yatvin. This defender of Whole Language reading teaching is a school principal from Boring, Oregon, a small suburb of Portland.
The NRP report is not credible, Yatvin unconvincingly contends, because the panel “has not fulfilled” its “obligation” to “settle the ‘Reading Wars’,” by “determining which of the many [reading] teaching methods used in schools,” that are “of the highest interest and controversy in the public arena,” are the ones that “really work best.” In short, the NRP report does not adequately “address the key issues” involved in the ongoing reading wars, Yatvin complains.
It is true, as Yatvin asserts, that the NRP report does not comment directly nor at length on the competing “theoretical models of reading” instruction. However, what the NRP did do, which is far more practical than to mull over theories, was to determine how closely the methods of instruction that these theories recommend conform to the available scientific evidence.
Particularly distressing to Yatvin in this respect are the NRP’s resolutions that direct and systematic instruction of children’s phonemic awareness, phonics skills, reading comprehension, and reading fluency are corroborated by experimental evidence. She also protests the importance that the panel places on children’s understanding of word meanings, ability to read words accurately and fluently, and reading to learn. By doing so, however, Yatvin reveals little more of significance than her exasperation with scientific findings on these matters.
In a further attempt to disparage the NRP’s report, Yatvin resorts to making accusations against it that have no foundation in fact. For example, the report does not testify in favor of separating “work pronunciation from work understanding.” Nor is it opposed to the development of students’ self-motivation to read, or to reading teachers “getting students to understand the main idea of a short story.” The report does not downgrade the need for students to “perform multiple [mental] operations in dealing with text,” and to use special “strategies in dealing with more difficult text,” Yatvin’s views to the contrary, notwithstanding.
Therefore, unwarranted is Yatvin’s pessimistic opinion that the NRP report, which “Congress intended to be a boon to the teaching of reading, will turn out to be a further detriment” to reform of this instruction. To the contrary, the panel clearly met its assignment to single out the most prominent aspects of children’s reading development, and to try to determine how experimental data indicate they are taught the most effectively. These issues are not “the only topics of importance in [children’s] learning to read,” the panel admits. But, the panel explains, the “sheer number of studies” on reading and its instruction “precluded an exhaustive analysis of the [experimental] research in all areas of potential interest.”
I predict that most Americans (and people from other English-speaking countries) who gain access to the NRP’s report will appreciate it. There thus should be widespread acceptance and application of the NRP’s recommendations in reading instruction programs in our nation’s public schools. Society must make sure that the changeover away from Whole Language teaching, that the panel’s report recommends, actually occurs. The demonstration of society’s civic responsibility in this regard will be the true test of the merit of the time, effort, and funds expended by the NRP to institute a return to scientific reading instruction.
Patrick Groff is professor of education emeritus at San Diego State University.