Evaluating Facts, Fictions, and Factions in the Reading Wars

Evaluating Facts, Fictions, and Factions in the Reading Wars

By Barbara W. Wise, Ph. D.

Current government policy requires that schools conform to the evidence-based recommendations of the Report of the National Reading Panel if they want federal funding for their reading programs (NRP: National Institute of Child Health and development NICHD, 2000; http://www.ed.gov/offices/OESE/esea/). These developments certainly improve the chances that scientific research may translate into powerful educational practice, especially if the government appropriates enough funds to allow schools to implement the policies well. We can thank Reid Lyon, the IDA, and other professionals and organizations for helping to move these policies along.

So, how are we doing? Do all our kids read their grade-level texts competently? Many reports suggest that average levels of reading have stayed relatively constant over the last 30 years (see Allington, 2002, p. 6-7). Sadly, the reading achievement of our poorest readers has not improved or has even declined in recent years (see Allington, 2002, p. 11-12; Simmons & Kameenui, 1998, p.3, and). However, particular studies and particular school districts are pointing the way toward every child becoming a reader (e.g., Grimes, 2002; Simmons & Kameenui, 1998; Torgesen, Wagner, & Rashotte, 1997).

Our San Diego symposium on Wednesday, November 12 is about myths and misconceptions. Under that topic, I could smugly consider old myths about reading

instruction, such as the misconception that “reading is (so) natural” that all kids should

get it as naturally as they learned to talk Goodman, 1976). Or I could debunk the myth

that any skills work is inherently and inevitably bad (Edelsky, 1990). But people in The

International Dyslexia Association (IDA) tend to know a fair amount about reading. I’d

rather not preach to the choir. So instead, I’ll look at fictions, questions, and controversies

relating to the NRP recommendations that are often tossed angrily about in battles about

Reading.

The NRP Report
The National Institute of Child Health and Development convened a large panel of researchers and educators to review scientific evidence in their Report of the National Reading Panel (NICHD, 2000). Those reviews led to policy recommendations that reading programs be “evidence-based” and cover five domains. The domains include a strong foundation in:

1) Phonological awareness and in

2) Phonics in reading and writing, and work on

3) Fluency,

4) Vocabulary, and

5) Comprehension.

These recommendations about domains appear strong and solid and align with the

practices endorsed by IDA and by many researchers and educators who recommend an informed and balanced approach to reading (Brady & Moats, 1997; Pressley, 1998). Of course, what that balance favors, when, and for which children is still under discussion. Hopefully, reading practitioners and program administrators will look at the evidence domains, consider how well their current practices cover them and adapt their practices or programs to cover them all well.

The report also suggests that all this instruction be intensive, systematic, and structured. Studies summarized in the report suggest that programs which use direct,

explicit, and systematic instruction in phonics and phonological awareness impact reading more powerfully than do programs that expect children to learn these things implicitly, embedded in stories, or taught in a non-systematic “as-needed” manner. But how explicit, for which students, for which aspects of reading, and for which teachers are questions still being debated (2002). The concept of explicitness is not yet clearly defined, research support is mixed, and more research is necessary to answer the questions precisely.

Consensus and Questions related to the Scientific Evidence
When I say consensus exists about evidence, I mean at least among the NRP report (2000), the report of the Research Council (NRC: Snow, Burns, & Griffin, 1998), and a recent excellent review by Rayner, Foorman, Pesetsky and Seidenburg (2002). When I say “strong consensus,” I also include a widely published author who raises questions and controversies about the NRP conclusions (Allington, 2002). Allington disagrees with the definition of evidence-base still thinks there is value in summaries of experimental evidence (p. 62).

A strong consensus exists that the strongest predictors of success in reading are phonological awareness and knowledge (see also Catts et al, 2001; Scarborough, 1998a &

b; Wagner, Torgesen & Rashotte, 1994). Phonological ability is at least partly due to

inherited factors (Gayán & Olson, 2000; Olson et al., 1989, 1999). The genetic pathways relate to brain-based differences that are described clearly and interestingly by Shaywitz

(2003). While a certain level of phonological awareness appears necessary to grasp the

alphabetic principle and learn to read (Liberman & Shankweiler, 1985), its relationship with reading is reciprocal (Perfetti, Beck, & Hughes, 1987).

That is, good phonological awareness improves reading, and it also improves with learning to read. The consensus about instruction is weaker. Systematic instruction in phonological awareness and phonics clearly benefits the reading accuracy of most children, and it can be taught in various ways (e.g., Torgesen, et al., 1997; Wise, Ring, & Olson, 1999, 2000). It helps level the playing field by improving the reading of at-risk children and poor readers, and the earlier it happens the better, in terms of increasing the number of children who will eventually reach grade level (Lyon, 1999; Torgesen, 2002). Early identification, screening, dynamic assessment in programs that include this instruction are cost effective and highly successful in some large-scale implementations (Grimes, 2002; Simmons & Kameenui, 1998). However, many researchers are not ready to say that evidence supports whether explicit phonics instruction is necessary for all children, nor do they agree on how much of it is needed (Allington, 2002; Olson, 2002).

If you inform yourself, you too can evaluate the evidence and the varying views of experts. I recommend you read the Rayner et al (2001) review mentioned earlier. If that looks daunting, the Scientific American published a much-simplified summary by the same authors in 2002. A raft of good recent review books can educate you further if you want (e.g., Bradley, Danielson, & Hallahan, 2002; Pressley, 1998; Simmons & Kameenui, 1998: Stanovich, 2000.) In evaluating talks, papers, and books, look for balance, clarity, and acknowledgement of weaknesses and remaining open questions. Listen to and read opposing views and keep their questions and challenges in mind as you evaluate research. The “Simple View” of reading can help you as you think about evidence (Gough & Tunmer, 1986). The model describes reading as an interaction among 1) all the processes that affect Word Reading, and 2) all the processes that affect Listening Comprehension. In evaluating the merits of methods in studies you read or hear about, consider how the method studied proposes to strengthen deficits and/or build on strengths as in this model.

As an educated consumer of research, watch that you don’t accept an evaluation or recommendation just because someone says, “Research says...” What research? How

was it done? The NRP and NRC reports and the Rayner review listed above discuss their

criteria for accepting studies for review and also describe the results behind their

recommendations. Remain skeptical until you can understand whether the study clearly

describes methods, student characteristics, interventions, and selection and training of

trained and control groups. If student success is the outcome of interest, then at the least,

teacher, student, and school variables should be measured and controlled. Trained groups

should be compared to untrained control groups or at least to standardized scores. If

children do change their standardized score ranks, it is as if they have improved relative

to a national control. But the most powerful studies for evaluating effectiveness of methods or programs compare two randomly assigned treatments, to control for the “Hawthorne” effects of an experimental group doing better merely due to the extra attention of being in a study.

Obviously, no study can do all things. Correlational and factor analysis studies

can allow us to see how variables interact in broad populations, but can’t tell us about

causation. Laboratory studies can control variables well, but have problems in generalization to the classroom situation. Classroom studies may generalize well, but

have less control. Our best shot at knowledge emerges from the converging evidence of

many varied studies.

In our Symposium about myths and misconceptions, I discuss and evaluate many

facts, fictions, and controversies that fuel the reading wars, in the hopes of improving our

understandings and improving the quality of the debate. But I’ll challenge you here with

one important fiction to consider right now, and you can look for the others in the symposium and the book that will follow in early 2004.

An important Fiction to Examine: We in IDA are on the “Right Side” in the Reading Wars

Even if you believe it, the danger is in the long-term outcome, of continuing the Wars at all. I vote for disarmament. We all may have heard some whole language advocate rudely “bash” one of our respected researchers or policy advocates. But what part do we play in continuing the wars: do we encourage any of this animosity with our own attitude of arrogance? Do we respect, listen to, and write about the knowledge that most teachers bring to the teaching situation, even as we hope that more of them will want to keep learning more and more about language and reading? Stanovich (2000), a scholar in our field, begs professionals and researchers on all sides to look for common ground and to debate civilly, in the interests of furthering knowledge. Lets consider the controversies and questions central to the battles and strive for ever -improving research and practice. If we hope to effect change, and help every child become a reader, then improving the quality of the debate is essential.

Right now, government policy encourages reading instruction and teacher education that is compatible with the mission of IDA. We in IDA have a chance now to help improve teacher education and implement strong programs. I hope we help it happen well, with an attitude of respect and inquiry. We can encourage schools and states to implement evidence-based programs and to document carefully what works and what doesn’t as these programs go forward. We can examine what doesn’t work, improve programs, and continue to encourage careful and creative research about reading and teaching. If we fail in this attempt and instead alienate more policy makers than we convince, we’ll have no one to blame but ourselves if the pendulum swings back.

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

Dr. Barbara Wise divides her work life among educational therapy, teacher education, and research. She has developed the Linguistic Remedies program for teachers, parents,

and children, with specific reading disabilities. She conducted research for 12 years using talking computers to study questions about reading remediation. Her current research is with the Center for Spoken Language Research at the University of Colorado.

She is helping her colleagues there to develop web-based interactive books and tutorial

activities, using animated, talking agents to improve children’s reading and to study

questions about reading remediation and its implementation with computers. Dr. Wise is

currently a member of the IDA Board of Directors and serves on several board

committees.

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

References and Footnotes Upon Request
Reprinted from the IDA 54th Annual

Conference Commemorative Booklet
“Our Mission to Literacy”