RIGHT AFTER becoming president, George W. Bush sent
Congress an education reform blueprint that included a proposal for "ensuring
that every child can read by the third grade" with a system of "scientifically
based reading instruction" based on the lessons of 100,000 research studies
reviewed by the National Reading Panel from 1967 to 2000.
In the versions of the education bill passed by the House and Senate and
soon to be finalized in conference committee, the term "scientifically based
reading instruction" appears a total of 52 times. Nobody is contesting it, but
they should be.
The reading education Bush has long had in mind can be gleaned from his
advice in 1996 that Texas teachers get "back to the basics" -that is, return to
traditional education that focuses on basic skills and facts. "We do not need
trendy new theories," Bush explained. "The basics work. If drill gets the job
done, then rote is right."
Although even the president's allies might take issue with the "rote"
comment, Bush accurately described the essentials of this teaching. "Trendy" is
code for reading instruction that stresses comprehension and teaches skills as
part of students' reading for meaning. "Basics" is code for beginning reading
instruction emphasizing explicit, direct and systematic instruction of skills
and minimizing the need for meaning and comprehension until skills are learned.
The trouble is, under such an approach, teachers follow established,
one-size-fits-all reading programs that move children through a step-wise
process from small parts of language to larger ones, with limited ability to
identify and address individual needs. A more precise reading of the research
evidence than Bush's shows that basics is hardly the only or best way for
children to learn.
The figure of 100,000 studies on reading is a number also cited by
Secretary of Education Ron Paige to justify the administration's program.
Unfortunately, this number does not remotely describe the actual number used in
the analysis by the National Reading Panel, which was created by Congress in
the second Clinton administration to assess the effectiveness of different
approaches to teaching children to read. (The reading researchers who advised
Bush as governor and now advise him as president are the same ones who selected
the members of the National Reading Panel in 1998.)
The panel did begin by identifying about 100,000 studies on reading but
pruned them extensively by narrowing the criteria-the studies had to be about
the "scientific" parts of instruction and so forth. At the end, there remained
52 studies on the ability to separate and manipulate speech sounds to identify
words; 38 studies for phonics; 14 for fluency; and 203 for comprehension.
Certainly these numbers are enough for drawing some reasonable conclusions, but
they do not add up anywhere close to 100,000. And they do not give enough
evidence to justify restrictive legislation mandating how all children should
What about the reading panel's findings themselves? The primary method of
evaluating the research was a meta-analysis, a statistical method that pools a
group of studies and estimates the average effect something has on something
else-in this case, the effect of aspects of reading instruction on student
achievement. While a meta-analysis can be useful, it can also have substantial
deficiencies. Such analysis will not, for example, distinguish the quality of
the reasoning used in the studies pooled. The reading panel's report is filled
with such problems almost entirely.
The panel describes, for instance, a study supposedly showing that children
trained to separate and manipulate speech sounds did much better than a group
without the training. (They learned letter names and sounds, were read to and
did some writing.) This provides evidence, wrote the panel, that the training
used in the study could be "used effectively in American classrooms."
This group comparison is correct as far as it goes, but the report does not
explain that there were two control groups-not one. The other group learned
skills "as needed"-when children demonstrated they needed a specific skill,
they were taught the skill directly and specifically. Research shows that
children learning to read this way tend to read more books, enjoy reading more,
are more thoughtful and critical readers, and write more. Teachers using this
approach tend to devote more time meeting individually with students to address
and assess their needs.
In this study, then, we have an opportunity to compare children learning
these skills either through a training program or an as-needed approach.
Although this comparison would have allowed the panel to delve into the
question of how these skills should be taught and learned, it did not do so.
Therefore, we will.
At the end of the school year, the skills-training group did significantly
better on tests requiring them to separate and manipulate sounds, but there
were "no significant differences between the groups on the tests of word
reading and spelling." The researchers had expected much higher reading scores
from the skills-trained group, but the National Reading Panel report did not
mention this finding.
The original researchers went on to propose that the "writing experiences"
of the skills-as-needed group might have accounted for their reading success.
On average, they "wrote longer stories than either" the training group or the
"normal" kindergarten group. These conclusions too were omitted. In other
words, the study showed that children can learn reading and spelling without a
rigid skills-training program. It suggests a more holistic approach, in which
skills are learned within a rich array of reading and writing activities, is
just as successful for teaching reading skills-with the added advantage of
giving the students more time learning how to write.
In "Misreading Reading: The Bad Science That Hurts Children," published
last year, I reviewed the large body of research that underpins claims about
"scientific" reading instruction and concluded that the studies flagrantly
failed to justify the step-wise reading instruction that now defines the Bush
reading legislation. Nothing in the National Reading Panel report refutes that
critique. Nor do three independent evaluations contradicting claims about the
so-called Texas "educational miracle" under then-Gov. Bush.
The Republican orchestration of hearings on the reading legislation
excluded all voices that could have offered an alternative perspective on the
claims of scientific evidence. Little wonder no one in Congress has raised even
the mildest objection to the proposed reading legislation. Little wonder that
no member of Congress seems to realize that this legislation will exclude
numerous effective instructional practices and diminish teacher initiative,
creativity and professional expertise.
One might reasonably conclude that having Congress legislate reading
instruction is itself dubious. But, given the lack of full information provided
Congress, a vote supporting federally crafted "scientific" standards for
reading instruction is misguided, unwarranted and even harmful.
In the days remaining before the final bill is drafted, the concept of
"scientifically based" instruction should be deleted from the legislation. This
will allow funding for other programs in which educators at the local level
can combine their study of reading research, knowledge of effective teaching
practices and understanding of students to arrive at the best way to teach
reading for each individual student.