NICHD Research: Setting the Record Straight

NRRF Note: The following review of Gerald Coles’ new book attacking NICHD and NICHD researchers was done by a respected non-NICHD researcher. The book, Misreading Reading: The Bad Science That Hurts Children, is being promoted widely and aggressively by the IRA, NCTE, and the WL communities. We are posting the review to set the record straight in an objective manner.

Straw Men and Very Misleading Reading: A Review of Misreading Reading
by Louise Spear-Swerling

Misreading Reading: The Bad Science That Hurts Children
by Gerald Coles
138 pp. Portsmouth, NH: Heinemann

When I was asked to review this book, I felt what might be described as guarded enthusiasm for the task. I had heard that Gerald Coles had written a new book on reading and I was curious to see it. On the one hand, even before I saw its subtitle, I knew that the book was probably going to express some views about reading quite different from my own. On the other hand, some of Coles’s earlier work, especially his 1987 book, The Learning Mystique: A Critical Look at “Learning Disabilities,” had influenced me a great deal. I found The Learning Mystique persuasive in many ways, and I believed that it made some legitimate criticisms of learning disabilities as a field of education. In Misreading Reading, I was anticipating a similar type of book—one that I might not entirely agree with, perhaps, but that would be substantial and thought-provoking.

OVERVIEW OF THE BOOK

Unfortunately, however, in both its writing and its scholarship, Misreading Reading differs greatly from The Learning Mystique. A central problem with Misreading Reading is its insistence on presenting the complex issues surrounding the field of reading primarily as a series of forced choices: systematic phonics or whole language, genes or environment, quantitative or qualitative research. In other words, the book is written as though only one choice of each alternative can be correct, when of course, this is not the case—nor is conceptualizing issues in this dichotomous way useful for understanding reading. Coles did not rely on this kind of thinking in The Learning Mystique, and I am puzzled that he does it here.

Misreading Reading focuses on a particular body of research in reading done by the investigators associated with the National Institute of Child Health and Human Development (NICHD). Each chapter of the book purports to disprove a particular claim, or set of claims, associated with code-emphasis or phonics viewpoints generally, and with certain NICHD studies in particular. The nearly exclusive emphasis of each chapter is on criticizing findings of specific studies. The book does not attempt to analyze anything approaching a total body of research on reading, or even on some specific area of reading, such as phonemic awareness. In addition, the book presents very little evidence in favor of an alternative viewpoint. Indeed, other than the fact that it looks favorably upon whole language instruction and considers children’s preschool literacy experiences important, the book does not say much about what such an alternative viewpoint might be. Space constraints apparently are the rationale for the lack of alternative viewpoints and evidence; in the introduction, Coles states that arguing “for or against the merits of whole language instruction and research . would require another book” (p. xvii). But the actual text of Misreading Reading is barely 100 pages long—surely there was room for something besides criticism of other people’s work.

An especially unpleasant aspect of the book is its emphasis on repeatedly discussing specific NICHD investigators by name. Ordinarily in scholarly reviews of research, the focus would be on criticizing ideas, not individuals, and individuals would be named primarily insofar as they are uniquely identified with a particular theory or argument (e.g., “Smith’s theory of sight word acquisition”). However, this book frequently calls into question not only the professional viewpoints, but also the personal integrity, of prominent NICHD researchers. The book does raise some legitimate issues about the translation of research findings into educational practice (e.g., when the researchers who are evaluating a program are the same people who developed that program, their ability to be objective may be compromised). However, the discussion of these issues is so one-sided, and so narrowly focused on criticizing a few specific individuals rather than on examining problems, that at times the book reads more like a celebrity gossip column than a professional or scholarly book.

Moreover, Coles often ascribes viewpoints to these researchers that they do not actually hold—the “straw men” of my title. Time after time when I looked up original sources reviewed in the book, I found that they had been discussed or quoted in a misleading manner. This misleading information is consistently at the expense of the NICHD researchers.

Overall, Misreading Reading is so obviously biased that it is tempting to dismiss it entirely and conclude my review here. However, I know that the book may seem persuasive to some readers, especially those not familiar with the original sources discussed in it. Therefore, I will spell out my objections, including specific examples of how the book misleads readers, in more detail below.

WHAT DO THE NICHD RESEARCHERS CLAIM?

Because their views are not always accurately represented in the book, it is useful to begin by discussing what the NICHD researchers actually claim (see, e.g., Adams, 1990; Blachman, 1997; Lyon, 1997). Although these researchers certainly are not identical in all of their viewpoints, in general they subscribe to what has been termed the “simple view of reading” (Gough & Tunmer, 1986). According to this view, two broad factors are responsible for individual differences in reading ability: skill at reading single words (word decoding), and general language comprehension (e.g., listening comprehension and vocabulary knowledge). Children who have difficulty learning to read can have difficulty in either area (word decoding or language comprehension), or in both areas.

However, especially in the early stages of learning to read, some children’s reading problems do revolve much more around word decoding than around general language comprehension; children classified as having learning disabilities typically fall into this group. Much research on reading, including NICHD-supported research, has focused on the abilities related to word decoding, especially phonemic awareness, or sensitivity to sounds in spoken words. Overall, the findings of this research support the viewpoint that the abilities involved in word decoding have a “necessary but not sufficient” relationship with reading comprehension. That is, good word decoding is a necessary foundation for comprehension, but other abilities—such as vocabulary, background knowledge, and the use of comprehension strategies—also are essential to good comprehension.

Research suggests that phonemic awareness has a reciprocal causal relationship with beginning reading; rudimentary levels of phonemic awareness clearly facilitate learning to read, but learning to read also increases phonemic awareness. (Reciprocal causation is typical of the relationship between many abilities and reading; for instance, vocabulary knowledge facilitates early reading, but learning to read also contributes to increases in vocabulary.) Explicit teaching of phonemic awareness, especially as part of a comprehensive program of reading instruction, appears to improve reading achievement. In addition, NICHD and other researchers have found that explicit teaching of word decoding skills (commonly termed phonics instruction) benefits beginning readers generally, and is particularly important to poor readers, who may require much more systematic and intensive instruction in this area than do other children. Again, these investigators would agree that phonics always should be part of a broader program of reading instruction.

Although some NICHD investigators have studied genetic influences on reading achievement, they acknowledge a wide variety of important environmental influences on reading as well. Just as poor readers can have difficulties that stem from a variety of cognitive weaknesses (e.g., poor phonemic awareness, poor word decoding, or poor general language comprehension), those cognitive weaknesses in turn can have a variety of ultimate causes, including lack of preschool literacy experiences, inadequate reading instruction, or hereditary factors.

Finally, many other lines of research concur with the basic NICHD findings. These include research from special education (e.g., research funded by OSEP, the Office of Special Education Programs); research conducted by CIERA (the Center for the Improvement of Early Reading Achievement); and research reviews done by a variety of scholars and scholarly panels (e.g., Adams, 1990; Anderson, Hiebert, Scott, & Wilkinson, 1985; Chall, 1983; Snow, Burns, & Griffin, 1998). For instance, a recently published CIERA study (Juel & Minden-Cupp, 1999-2000) examined the reading instruction of four experienced first-grade teachers at two demographically similar schools. The study tried to determine how classroom instruction affected children’s growth in reading. Children who were at greatest risk in reading—those with the greatest weaknesses in beginning literacy skills—did best with the teacher who employed the most structured, systematic teaching of phonics for the first half of first grade.

Of course, different researchers have frequently chosen to focus on a specific area of interest in their work. And different researchers certainly would disagree about many specific issues that I have not discussed here. But none of them would make simplistic claims about one best program for teaching reading, one cause of reading failure, or one solution to reading problems.

CONTENTIONS OF THE BOOK

Now consider the “claims” of NICHD researchers as presented in Misreading Reading:

  1. “Phonemic awareness (hearing, distinguishing, and manipulating the sounds in words) is the chief causal factor in early reading achievement and the ‘core deficit’ in reading problems.” (p. xx)
  2. “Difficulty understanding that words are made up of sounds and learning the ‘alphabetic principle’ (associating sounds with alphabet letters) are the primary causes of poor reading.” (p. 16)
  3. “NICHD-supported research has identified a ‘brain glitch’ that causes reading disabilities (also known as dyslexia).” (p. 60)
  4. “Genes have been identified that can cause reading disabilities.” (p. 70)
  5. “As if by magic, many children acquire phonological awareness and rudimentary literacy abilities prior to beginning schooling. Those children who have not . require direct, systematic instruction in reading and writing skills.” (p. 90)
  6. “Scientific evidence demonstrates the need for direct, explicit instruction of ‘alphabetic code’ skills in beginning reading and the likelihood that this kind of instruction is the most powerful weapon for promoting literacy.” (p. 100).

These supposed claims, however, are substantially stronger than the ones that most NICHD researchers would make, and they lack the qualifiers that typically characterize scientific claims. For instance, most NICHD investigators would not say that phonemic awareness is “the chief causal factor in reading achievement,” although they would certainly say that it is an important causal factor. Also, they would acknowledge the complexity of the causal relationship between phonemic awareness and reading achievement, which is bidirectional. Similarly, most researchers would not argue that a “brain glitch” or genes, by themselves, “cause reading disabilities,” although many would say that biological factors predispose certain individuals to reading disabilities. These distinctions are important, because they illustrate the tendency of Misreading Reading to cast complex issues as forced choices: reading disabilities must be biological or environmental, not both; phonemic awareness causes or is caused by early reading achievement, not both. As I have already noted, this forced-choice format is not a very useful way to think about reading and reading problems.

In addition to the discussion of the preceding “claims,” Coles strongly criticizes two published programs, the Open Court reading program, employed in an intervention study conducted by Barbara Foorman and her colleagues (Foorman, Francis, Fletcher, & Schatschneider, 1998), and Phonemic Awareness in Young Children: A Classroom Curriculum (Adams, Foorman, Lundberg, & Beeler, 1998). Many of Coles’s criticisms of these and other, similar programs (e.g., in Chapter 8 of Misreading Reading) revolve around three specific charges. First, he claims that inadequate comparison groups often have been used in the studies cited as support for the programs; for example, groups designated as receiving “whole language” or meaning-emphasis instruction actually were not receiving such instruction. Second, according to Coles, programs such as Open Court only improve word-level reading skills, not comprehension. And third, he says, these programs do not bring about long-lasting effects on reading achievement. I will briefly consider each of these criticisms.

With regard to the first charge, it is true that the nature of the comparison groups is important in intervention studies; for instance, if the instruction in the comparison group is very poor, almost any intervention may look good by comparison. Many intervention studies, including those criticized in Misreading Reading, have used as at least one comparison group the regular instruction typically delivered to children in a particular school. Because of the popularity of meaning-emphasis and whole language instruction in the United States, often this means instruction that the teachers themselves describe as “whole language instruction.” Coles repeatedly argues that the instruction provided in various studies is not “good” whole language instruction, but he offers no specific guidelines as to exactly what constitutes good whole language instruction—nor any evidence to show that such instruction is more effective for any group of children than are programs like Open Court.

Second, although researchers sometimes have had difficulty demonstrating that the benefits of phonemic-awareness and word-decoding instruction transfer to reading comprehension, even in studies without statistically significant findings for comprehension, there frequently is a trend toward improvement in comprehension. Because good reading comprehension involves many abilities, some children may improve word decoding without showing a concomitant increase in comprehension. Coles repeatedly criticizes the interventions of different investigators because they “only” improve word decoding, but evidence strongly suggests that good out-of-context word decoding is not a trivial skill. For instance, individual differences in word decoding are strongly associated with individual differences in reading comprehension. It is unusual for elementary children’s reading comprehension to exceed the level that would be expected by their ability to decode words, and even more unusual for their comprehension to be below their level of decoding (Shankweiler et al., 1999). Coles’s criticisms are not a good argument for eschewing phonics instruction; rather, they suggest that there must be more to reading instruction than just phonics, a point with which the NICHD researchers would agree.

Finally, Coles is correct that programs like Open Court, Phonemic Awareness in Young Children, and other phonics-based interventions, by themselves, do not provide a lasting “cure” of poor reading for all children. As I have already indicated, NICHD researchers would not make such a claim. Demonstrating long-term benefits has been problematic for many intervention programs in reading (see, e.g., Hiebert’s 1994 critique of Reading Recovery). It certainly is possible that there are other, as yet untested, ways of intervening with poor readers that would enhance long-term effects, and research in this area is ongoing. However, criticism of existing studies is not a substitute for providing alternative evidence of a more effective way to intervene.

POLITICS AND READING

According to Misreading Reading, all of this “bad science” done by NICHD investigators would not be so terrible if it were not part of “a seamless connection between research, policy, and politics” (p. 108). Specifically, Coles charges, NICHD research “is part of a full-scale effort to extinguish whole language from the schools” (p. 103); excludes important influences on literacy achievement, such as poverty, literacy in the home, and children’s interests; and limits definitions of scientific knowledge to certain kinds of studies (i.e., controlled experimental studies employing quantitative data).

However, Misreading Reading itself has an obvious political agenda. This agenda is evident on the back cover of the book, which describes Misreading Reading as “an opportunity to enact early reading policy and legislation based on ‘good science,'” and which suggests that the “reader-friendly summaries” in the book “can serve many possible uses—including ‘talking points,’ handouts, or overheads for presentations.”

An examination of how political influences shape reading instruction and policy is a worthy topic for a book. But to be credible, such a book must recognize that political influences operate on all sides of the reading field, not just on one side. An effective book on this topic would focus on problems and issues rather than attacks on a particular group of people; would recognize a variety of motivations in those who have tried to shape reading policy, including altruistic motives; would avoid presenting complex issues in dichotomous terms; and would impartially review a wide range of evidence. Misreading Reading has none of these characteristics.

For example, there are many good qualitative and ethnographic studies (e.g., Heath, 1983) with implications for understanding reading difficulties. These qualitative studies might be examined in conjunction with the NICHD research—not by setting quantitative and qualitative research in opposition to each other, but by considering how different types of studies may offer insights about different kinds of variables important in literacy acquisition. (For a good example of this approach, see Adams, 1990.) Of course, both qualitative and quantitative studies should adhere to certain basic principles of science, including gathering evidence through systematic observation and testing, stating claims in ways that are potentially falsifiable, considering alternative explanations of evidence, and considering a cumulative body of evidence. However, not only are guidelines for determining what constitutes “good science” absent from Misreading Reading, so is a discussion of most qualitative and ethnographic studies relevant to reading. As an example of an alternative scientific methodology, Coles offers an anthropological study of menstruation in West Africa (p. 107).

NICHD research does not have as its goal “extinguishing whole language from the schools.” Certainly the findings of this research are at odds with some instructional practices commonly associated with whole language, such as the idea that highly contextualized phonics instruction is effective for all children. However, NICHD research actually supports other practices advocated by whole language proponents, such as their emphasis on reading to children and providing them with abundant exposure to text. A more fruitful way to examine these issues is not in terms of the tired “phonics versus whole language” debate, but in terms of which specific instructional practices are most useful with which kinds of children, at which stages of development. An excellent example of this kind of approach is provided by Brown (1999-2000), who discusses how predictable text, decodable text, and authentic children’s literature all can be useful in scaffolding instruction for children who are at different stages of reading development.

Finally, Coles makes a legitimate point when he argues that we need to look at myriad influences on children’s literacy learning, not just at a narrow set of variables such as phonemic awareness or word decoding. But reading research and policy should not pit these influences against each other; the study of word decoding, for instance, need not detract from studying (and addressing) broad social factors in literacy learning. However, it is hardly surprising that some educators and educational psychologists have chosen to focus on the one variable over which teachers can exert the most control—the nature of reading instruction.

MISLEADING USE OF SOURCES

In my overview of the book, I noted that the original sources for Misreading Reading were sometimes quoted or discussed in misleading ways. Before concluding my review, I would like to give some specific examples of what I mean, especially for potential readers of the book who may not be familiar with the studies reviewed in it. Although it was not feasible for me to investigate every source cited in the book, in the many I checked, I found numerous instances of sources that were used in a misleading manner.

One example of a misleading quotation involves Reid Lyon’s 1997 testimony before Congress. At the beginning of his testimony, Lyon uses the phrase “almost as though by magic” to refer to the apparent ease with which some children learn to read. Coles highlights this phrase (which is also included in one of the “claims” he attributes to NICHD researchers) to suggest that Lyon and other NICHD investigators view early reading acquisition as some sort of mysterious black box: “[The phrase ‘as though by magic’] gives the impression that this blossoming cannot or need not be explained by evidence and additional examination, when in fact there is a considerable body of evidence explaining the preschool influences on reading progress that might seem magical to Lyon and others” (Coles, 2000, p. 91).

However, one has only to read further in Lyon’s testimony to see that Lyon not only acknowledges, but actually emphasizes, preschool experiences as important causal influences in beginning reading achievement—although he also suggests that these experiences, or the lack of them, cannot account for all cases of reading failure: Difficulties learning to read result from a combination of factors. In general, children who are most at risk for reading failure are those who enter school with limited exposure to language and thus less prior knowledge of concepts related to phonemic sensitivity, letter knowledge, print awareness, the purposes of reading, and general verbal skills, including vocabulary. Children raised in poverty, youngsters with limited proficiency in English, children with speech and hearing impairments, and children from homes where the parent’s reading levels are low are clearly at increased risk of reading failure.. However, it is very important to note that a substantial number of children from highly literate households and who have been read to by their parents since very early in life also have difficulties learning to read. (1997, p. 6)

Moreover, in his recommendations for helping children learn to read, Lyon places preschool experiences at the top of his list: .Success in learning to read is based in large part on developing language and literacy-related skills very early in life. A massive effort needs to be undertaken to inform parents, and the educational and medical communities, of the need to involve children in reading from the first days of life; to engage children in playing with language through nursery rhymes, storybooks, and writing activities; and, as early as possible, to bring to children experiences that help them understand the purposes of reading, and the wonder and joy that can be derived from it. (1997, p. 10)

These sections of Lyon’s testimony, however, are not discussed in Misreading Reading.

A second example of a misleading quotation is provided by Coles’s discussion of a study conducted by Hatcher, Hulme, and Ellis (1994). Hatcher and his colleagues studied the effectiveness of four different interventions with a group of 7-year-old poor readers: one in which phonological abilities were trained in isolation from reading instruction, one in which reading but not phonological skills was taught, one in which phonological skills and reading were taught together, and a control group. Not surprisingly, the most effective intervention for enhancing reading achievement proved to be the one that taught phonological skills and reading in conjunction. Hatcher et al. summarize their findings in a paragraph that is quoted in Misreading Reading (p. 86); however, the final sentence, which I have italicized, is omitted by Coles: These findings therefore cast doubt on the simple theory that there is a direct causal path from phonological skills to reading skills. Our data support the more subtle position that adequate phonological skills may be necessary, but not sufficient, for learning to read effectively. (Hatcher et al., 1994, p. 53)

As I have already explained, the “necessary but not sufficient” view is the one that would be endorsed by NICHD investigators.

Another interesting aspect of this study is that Hatcher et al. (1994) used an adaptation of Clay’s (1985) Reading Recovery program as the foundation for their “Reading Plus Phonology” intervention. Hatcher and his colleagues modified the Reading Recovery program primarily by adding more systematic phonological instruction to it. As Hatcher et al. note, their results are consistent with those of other studies (e.g., Iversen & Tunmer, 1993) which have found that the effectiveness of Reading Recovery is enhanced by the addition of a component involving systematic phonological training. However, this finding of Hatcher et al. (1994) is not mentioned inMisreading Reading; neither is Iverson and Tunmer’s work.

My final example of how this book misleads readers involves Coles’s review of a widely cited paper by Wagner and Torgesen (1987). As Coles notes, these researchers reexamined the data of another group of investigators (Lundberg, Olofsson, & Wall, 1980), who had found that Swedish kindergartners’ phonemic awareness predicted their later reading and spelling achievement. Wagner and Torgesen’s criticisms of the Lundberg study revolved around the fact that Lundberg and his colleagues did not include in their analyses measures of kindergarten reading achievement (e.g., knowledge of letter names and sounds). That is, Wagner and Torgesen pointed out a potential confound in the Swedish study. Because the best predictor of future performance is often present performance, kindergarten reading achievement, rather than phonemic awareness, might have been the important variable in predicting children’s later reading achievement. Wagner and Torgesen’s criticism of the Lundberg study is quoted in Misreading Reading without further explanation, as part of a critique of Phonemic Awareness in Young Children: A Classroom Curriculum (Adams et al., 1998) in Chapter 4.

What is not discussed, however, is a later longitudinal study by Wagner, Torgesen, and Rashotte (1994) involving a group of 244 children followed from kindergarten through second grade. This later study addressed the confounds in the Lundberg work and helped to clarify the causal relationship between phonological abilities and beginning reading. The findings of Wagner et al. (1994) support the viewpoint that, although there clearly is a reciprocal relationship between phonological abilities and early reading, phonological abilities do play an important causal role in beginning reading achievement. Specifically, phonological abilities (including phonemic awareness) exerted a strong causal influence on word decoding; knowledge of letter names (but not word decoding) exerted a more modest causal influence on subsequent phonological abilities.

Wagner et al. also found that individual differences in phonological abilities were remarkably stable throughout children’s first few years in school. Of course, individual differences in kindergartners’ phonological abilities might be related to individual differences in preschool experiences with literacy, or to some other experiential variable, rather than to innate differences in ability. However, the value of phonemic awareness and phonics instruction does not hinge on the ultimate cause of individual differences in phonological abilities. Similarly, for instance, children who enter school with vocabulary weaknesses may benefit from vocabulary instruction regardless of whether their weaknesses stem from lack of innate ability or from lack of exposure to words.

Many other original sources in Misreading Reading are discussed in misleading ways, including Stanovich’s 1986 paper on “Matthew Effects” in reading, Foorman et al. (1998), and Adams (1990). This misleading use of sources in a manner that always disfavors the same viewpoint—that of the NICHD researchers—greatly undermines the credibility of the book as a whole.

CONCLUSIONS

It certainly would be possible to write a constructive, useful critique of reading research, including NICHD research, and how it may be misapplied in educational practice. However, a useful critique would represent different researchers’ viewpoints fairly, review evidence related to alternative viewpoints impartially, and acknowledge that political influences operate across all viewpoints in the reading field. A thoughtful book might address some of the following kinds of questions: Exactly what constitutes “good science”? How can we avoid having research findings interpreted in an overly simplistic fashion, whether those findings pertain to phonics instruction or exposing children to good books? What kinds of political influences, across a variety of viewpoints and issues, affect reading instruction? What does research suggest about some ways for teachers to meet the needs of diverse groups of children, whether those children vary due to differences in ethnicity, culture, economic background, or native abilities? Misreading Reading does not even ask any of these questions, let alone provide answers to them. It’s too busy finding fault with the NICHD investigators.

Perhaps the most disturbing aspect of this book is that it represents more of exactly what the reading field does not need—the same old “good guys/bad guys” approach, the same old polarization of reading professionals into opposite camps, the same old simplistic ways of framing complex issues. Misreading Reading certainly is not the only volume to have used these tactics, which have been employed on both sides of the phonics debate. But books like these have never provided fresh insights about, or constructive solutions to, the complex problem of reading failure.

References

  • Adams, M. J. (1990). Beginning to read: Thinking and learning about print. Cambridge, MA: MIT Press.
  • Adams, M. J., Foorman, B. R., Lundberg, I., & Beeler, T. (1998). Phonemic awareness in young children: A classroom curriculum. Baltimore, MD: Brookes.
  • Anderson, R., Hiebert, E., Scott, J., & Wilkinson, I. (1985). Becoming a nation of readers: The report of the Commission on Reading. Champaign, IL: Center for the Study of Reading.
  • Blachman, B. (1997). Foundations of reading acquisition and dyslexia: Implications for early intervention. Mahwah, NJ: Erlbaum.
  • Brown, K. J. (1999-2000). What kind of text—for whom and when? Textual scaffolding for beginning readers. The Reading Teacher, 53, 292-307.
  • Chall, J. (1983). Learning to read: The great debate (revised). New York: McGraw-Hill.
  • Clay, M. M. (1985). The early detection of reading difficulties (3rd edition). Portsmouth, NH: Heinemann.
  • Coles, G. S. (1987). The learning mystique: A critical look at “learning disabilities.” New York: Pantheon Books.
  • Foorman, B. R., Francis, D. J., Fletcher, J. M., & Schatschneider, C. (1998). The role of instruction in learning to read: Preventing reading failure in at-risk children. Journal of Educational Psychology, 90, 37-55.
  • Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading disability. Remedial and special education, 7, 6-10.
  • Hatcher, P. J., Hulme, C., & Ellis, A. W. (1994). Ameliorating early reading failure by integrating the teaching of reading and phonological skills. Child Development, 65, 41-57.
  • Heath, S. B. (1983). Ways with words. Cambridge: Cambridge University Press.
  • Hiebert, E. H. (1994). Reading Recovery in the United States: What difference does it make to an age cohort? Educational Researcher, 23, 15-25.
  • Iverson, S., & Tunmer, W. E. (1993). Phonological processing and the Reading Recovery Program. Journal of Educational Psychology, 85, 112-126.
  • Juel, C., & Minden-Cupp, C. (1999-2000). One down and 80,000 to go: Word recognition instruction in the primary grades. The Reading Teacher, 53, 332-335.
  • Lundberg, I., Olofsson, A., & Wall, S. (1980). Reading and spelling skills in the first school years predicted from phonemic awareness skills in kindergarten. Scandinavian Journal of Psychology, 21, 159-173.
  • Lyon, G. R. (1997). Report on learning disabilities research. From testimony of G. Reid Lyon on children’s literacy. Committee on Education and the Workforce, U. S. House of Representatives, Washington D. C. Available at www.ldonline.org.
  • Shankweiler, D., Lundquist, E., Katz, L., Stuebing, K., Fletcher, J., Brady, S., Fowler, A., Dreyer, L., Marchione, K., Shaywitz, S., & Shaywitz, B. (1999). Comprehension and decoding: Patterns of association in children with reading difficulties. Scientific Studies of Reading, 3, 69-94.
  • Snow, C. E., Burns, S., & Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press.
  • Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21, 360-406.
  • Wagner, R. K., & Torgesen, J. K. (1987). The nature of phonological processing and its causal role in the acquisition of reading skills. Psychological Bulletin, 101, 192-212.
  • Wagner, R. K., Torgesen, J. K., & Rashotte, C. A. (1994). Development of reading-related phonological processing abilities: New evidence of bidirectional causality from a latent variable longitudinal study. Developmental Psychology, 30, 73-87.