One of the most audacious promises of the landmark No Child Left Behind Act of 2001 was its call for every child in the United States to be proficient in reading and math by the year 2014. This proficiency target, coupled with the disaggregation of test scores by traditionally underperforming groups, such as African-American and Hispanic students, drove the education system for the last decade. A political compromise at the time has yielded a system that works against the interests of students, parents, and teachers.

Given the historical context in which the Bush administration was operating in 2001—especially the bloody fight over a voluntary national test that marked the last years of the Clinton administration—the federal government was able to establish 100 percent proficiency as the national goal only by giving the states the authority to define proficiency and develop the tests to assess student competence.

The result was not surprising: states created low proficiency standards. The most systematic comparison of state proficiency standards has been conducted by the National Center for Education Statistics (NCES), which mapped state proficiency “cut-points” onto those established by the National Assessment of Education Progress (NAEP), often called the nation’s report card. NCES has reported on state vs. NAEP proficiency cut points in 2007 and again in 2009. Taking fourth-grade reading as an example of just how low state-proficiency requirements are, the 2009 report found that 31 states set scores for reading proficiency below NAEP’s Basic level and no state set its proficiency score equal to NAEP’s.

Different Worlds

The implications of this broken system of testing are nowhere more apparent than in the separate reports that NCES produces focusing on a number of large urban districts. Called the “Trial Urban District Assessment” (or TUDA), the reading results for students in 18 large urban districts in fourth and eighth grades were released in mid-May (the math results were released in December 2009).

The students in these urban districts are predominately from racial and ethnic minorities and disproportionately from low-income families—the groups where educational challenges are greatest and where measuring performance and growth most critical. The TUDA results from the 2009 NAEP reading assessment gives further insight into how compromised the system of testing we created in 2001 is. Below, I discuss the fourth-grade reading results—but the eighth-grade results are virtually the same.

In figure 1, we see that state reading assessments and NAEP describe two different worlds of student achievement. Across these large urban districts, according to state tests, 56 percent of fourth graders are reading at a proficient level—this is almost three times higher than the percent of fourth graders meeting NAEP proficiency standards (20 percent).

In Atlanta, Baltimore, and Houston, the difference between the state and the NAEP percent proficient is greater than 60 percentage points, and in Austin and Detroit it is more than 50 percentage points. Boston’s percent proficient comes close to NAEP’s—and this is the result of a long-standing effort in Massachusetts to have rigorous standards and assessments.

It is true that NAEP has rigorous standards and that it has aspirational goals in setting its cut-point for proficiency. And it is also true that state assessments serve somewhat different functions than NAEP. But can these factors explain the average gap of over 40 percentage points between what the states and NAEP say are the percent of students in these urban districts that are proficient?

Education reforms during the last couple of decades have been aimed at making progress, and NCLB codified that goal. Figure 2 shows how students in these districts progressed on state tests and on NAEP. Eleven districts had data on both state assessments and NAEP in 2007 and 2009, and I show the difference in percent proficient in each of these districts on the two assessments.

The “good news” is that in seven of the eleven districts, at least the direction of the change was the same on the two tests. In Chicago and Los Angeles, the state tests showed gains (in Los Angeles a jump of over 10 percentage points), while NAEP reported no change. In Charlotte and Boston, the percent proficient declined on the state test while their fourth-grade students registered gains in NAEP.

These discrepancies between NAEP and state assessments are no secret—indeed, Arne Duncan, the secretary of Education, has gone so far as to say that states are “lying” to parents and students about their performance. Clearly, the compromise made in 2001—allowing states to define proficiency—hasn’t worked.

As the Congress begins to gear up for reauthorizing the Elementary and Secondary Education Act, it must gut this component of NCLB. In the meantime, the move toward national standards and assessments based on those standards continues to gain steam, and even the ever-skeptical Checker Finn has judged the initial work “unexpectedly and encouragingly good.”

Perhaps this time around, the nation can become more proficient in defining proficiency.

Mark Schneider is a visiting scholar at the American Enterprise Institute and vice president at the American Institutes for Research.

FURTHER READING: Schneider discussed “How Bad Are Our Graduation Rates?” and said the No Child Left Behind law means “We’re Still ‘Lying to Our Children.’” He also described why “Obama’s Education Hopes Face Achievement Realities,” documented with other scholars how some institutions are “Rising to the Challenge” of raising Hispanic college graduation rates, and explained “Where Does All That Tuition Go?”

Image by Darren Wamboldt/Bergman Group.