|
|
|
Companies, People, Ideas
Children Will Be Left Behind
03.15.04
George Bush's school accountability law,
enacted to much fanfare two years ago, is something of a fraud. It cannot
possibly perform as advertised.
The No Child Left Behind law, which sailed through Congress
with overwhelming majorities two years ago, has a giant problem--one that will
cause the act to fail. But no one discusses this problem in public.
Even the law's fiercest critics--who now include just
about all our country's prominent Democrats--seem not to have noticed the real
problem. And it certainly will not be pointed up by such longtime enthusiasts
as the Business Roundtable, the National Association of Manufacturers, the U.S.
Chamber of Commerce and numerous high-profile chief executives. They like the
"accountability" the law promises to deliver. They like its incentive
system, which steers funding to successful schools (as measured by test scores)
and penalizes the failures. They like the higher standards for teachers, and
the threat these pose to the teachers' unions. They are even learning to love
the U.S. Department of Education, which now spends $55.6 billion a year
administering No Child and other federal programs, and they are presumably
pleased that Ronald Reagan welshed on his 1980 campaign promise to ax the
agency.
Last year was the first in which the entire No Child
machinery was up and running, and we learned a few things about how it will
work. The then-current crop of news stories reflects the exasperation of local
school officials, who did not expect so much paperwork and gripe about
"unfunded mandates." Another familiar story line centers on the shock
of administrators at first-rate schools when told they are "failing"
(or at least that term keeps getting into the headlines).
The alleged failure often involves technicalities. No
Child's authors were determined to forestall cheating by principals, many of
whom had long boosted their schools' test scores by encouraging poor students
to stay home on days when big tests were given. So the new law provided that
95% of all students--and in some cases 95% of each ethnic group within the
school--had to participate. Inevitably, some schools were flunked because they
only had, say, 94.6%. As this article goes to press, several states--Virginia,
Minnesota and Utah among them--look like they might opt out of No Child, as the
law allows them to do. Any such decision would mean a loss of some federal
funding for education but would lift the new regulatory burden.
And yet the law's main problem continues to be
unrepresented in the news stories. The problem is that some students are not
smart enough to do well on tests. This might be considered too obvious to
mention but for some astounding details about No Child. For openers, it
proposes to eliminate--not reduce, eliminate--the "achievement gap"
between prosperous and impoverished students. The gap is tremendous and in
large measure reflects socioeconomic IQ differences. The states with the most
students eligible for the federal free/reduced lunch program (a fairly good
indicator of poverty status) reliably produce the lowest reading and math
scores.
But No Child's IQ problem is not just a matter of
social class differences. The law also states, insanely, that by 2014 all
American students must be "proficient" in reading and math. Any
school at which this doesn't happen will suffer severe penalties, up to and
including a takeover by the state. Yet the shape of the bell curve guarantees
that most schools will fail. No amount of accountability, incentives and
superduper teaching can possibly get all the kids in any sizable school up to
100% proficiency by 2014. The act supported by all those hardheaded businessmen
is utterly utopian.
To be sure, 2014 is ten years off. But during those
years each state must continually demonstrate that it is making "adequate
yearly progress" at a rate that will take it from present levels of
proficiency to 100% by 2014. And the yearly progress requirements have had
perverse effects in many school districts. In some states the effect has been
to lower academic standards.
No Child is lowering standards? How can that be? The answer
resides in the fact that in order to make the new law seem manageable its
authors gave the states some wiggle room in defining "proficient." No
Child envisions four levels of mastery for each subject: "Advanced"
is highest, followed by "proficient," "basic," and (the
lower depths) "below basic." This four-tiered schema was copied from
the National Assessment of Educational Progress, which has regularly surveyed
the academic achievements of American kids since 1969. But--critical
detail--the states were not required to embrace the NAEP definitions of those
terms. In the NAEP tests, proficiency is defined as "solid academic
performance … demonstrated competency over challenging subject matter … and
analytic skills appropriate to the subject matter."
The definition is, to be sure, fuzzy, but it comes
accompanied by some test questions that make it real and that guarantee most
students will be nowhere near proficient. In the 2003 NAEP sample only 27% of
the country's eighth graders were proficient or better in math; 33% were below
basic. So, looking at their new situation, the states decided overwhelmingly
not to go for the NAEP standards.
But that was not all: It also sank in that their yearly
progress burdens would be lower if their standards were lower. A lower standard
means that a higher proportion of students will already be close to proficient,
which means in turn that the required annual progress will be less demanding.
Only three states (Louisiana, Delaware and Connecticut) opted for the NAEP standards.
All the others understandably decided either to stick with their traditional
standards, which were lower than NAEP's, or to reduce their standards further.
I spoke recently with Sharif Shakrani, deputy executive director of the
National Assessment Governing Board (the policy board that runs NAEP), who said
crisply: "I can understand that. If the states make [the standards]
tighter, they will just have more that are failing."
But holding down standards does not entirely solve the
states' problems with No Child. For one thing, there are limits to how low you
can go. If standards begin to seem a travesty, you get howls from the parents.
And with any kind of meaningful standards at all, 100% proficiency is
impossible. Robert Linn, the new president of UCLA's respected National Center
for Research on Evaluation, Standards and Student Testing, recently observed
that it would be an enormous challenge just getting 100% of kids to
NAEP's"basic" level.
Nobody can say for sure how the drama will play out, but
one way or another, No Child will be changed. Its goals are wildly unrealistic,
and a sizable fraction of the educators now caught up in the process know it's
unrealistic. As the late economist Herbert Stein famously said, "If
something can't go on forever, it will stop." He was talking about
financial deficits, but he might as well have been talking about deficient
thinking in educational reform.