What Really Matters for Struggling Readers: Designing Research-Based Programs (What Really Matters Series)

by Richard L. Allington

The report card details the findings of the National Assessment of Educational Progress (NAEP), a series of assessments covering an array of subject areas. The NAEP reading assessments are the most frequently administered and seem to generate the most discussion. The general theme of such discussions recently has gone something like this: “We’ve dramatically increased education expenditures, but reading scores remain flat with huge achievement gaps between different subgroups.”

TABLE 1.1 Combined Reading Literacy Scores, Ages 9 and 15

FIGURE 1.1 NAEP Reading Long-Term Trend Scores for the Nation

FIGURE 1.2 NAEP Main Report Reading Scores 1992–2009

The achievement of U.S. elementary and middle school students on nationally normed, standardized commercial tests of reading achievement have been rising since 1980.

Finally, one other indicator that might be used is the readability formulas that were created to estimate the difficulty of books.

The first is the trend for certain groups of children to lag behind their peers in literacy learning.

They concluded that only parent educational levels and family income were related to achievement.

Being poor places you at risk of reading difficulties; being a boy or being a minority also places you at risk.

These are the children of low-income families where levels of parental educational attainment are also typically low. If our schools fail to teach these children to read well, how likely is it they will ever become productive citizens?

We seem to be producing readers who can read more difficult texts but who elect not to read even easy texts on their own time.

The point to be made is that being able to locate or remember the correct answer (word or phrase) to a multiple-choice item, the traditional measure of comprehension, is simply much less demanding than responding to these new measures of reading comprehension.

Similar findings after 27 coordinated studies compared methods and materials in the largest beginning reading field experiment ever conducted led the authors of the First Grade Studies (Bond & Dykstra, 1967) to conclude: “Children learn to read by a variety of materials and methods. . . . No one approach is so distinctly better in all situations and respects than the others that it should be considered the one best method” (p. 75).

Unfortunately, rather few studies report effects over periods greater than one year, and many report the effects after only a few months.

“The initial hypothesis, that by adopting a whole-school design a school could improve its performance was largely unproved” (Berends et al., 2002, p. 173).

In most cases, the available studies evaluated the effects of a literacy intervention on assessments of basic literacy, not on thoughtful literacy assessments. In other words, most studies use word lists, tests of subskill knowledge, or assessments of low-level comprehension found on traditional standardized tests with their multiple-choice items.

Elementary and Secondary Education Act of 1966

A key component of ESEA was Title I of that act. Title I focused on providing high-poverty schools with additional funds to support supplementary reading instruction.

However, in far too many states paraprofessionals or teachers with no graduate training in reading disabilities provided the extra lessons. In far too many schools paraprofessionals simply monitored struggling readers while they worked on some form of computer-assisted reading lessons.

That is why I noted on page 1 of this book that I was not surprised that the Reading First initiative failed to improve reading achievement (Allington, 2009b).

Reading instruction in NCLB schools, at least, was to be based on “scientifically based reliable, replicable research,” or SBRR. Unfortunately, little research was actually consulted when designing NCLB or in designing the reading instruction under NCLB.

Two problems existed that the NCLB designers ignored. First, there was no research supporting the use of any of the core reading programs and none that supported the use of any supplementary reading program, except Reading Recovery.

Second, what the research seems to suggest is that adaptation of commercial reading programs, adaptations based on student needs, produces better results than just following the reading program guidelines. But such adaptations are made only by teachers who are reasonably expert about b...

doing so under the guise of “scientifically based reliable, replicable research.” Congress has terminated funding for the Reading First portion of Title I funding, at least in part because the large federal study of the effects of the Reading First program indicated that no reading achievement gains occurred in Reading First schools

The “corruption” involved recommendations for programs and assessments that had no research base, but these programs and assessments did have authors who were involved in making the decisions about what programs and assessments should be used.

Adequate Yearly P...