More Confusion on NCLB Scores
Ohanian Comment: This article says NCLB make it difficult to compare schools in different states. Why does anyone besides a newspaper reporter need to compare Maryland schools with Virginia schools? Most real estate agents don't even see a need to do this.
More to the point is the "race to the bottom."
Deep Run Elementary School in Howard County almost made it.
Nearly every group of students at Deep Run -- white children, black children, Hispanics, even those who speak little English -- scored well enough on Maryland's new standardized tests this year to meet the requirements of the federal No Child Left Behind law. But because all 10 special-education students who took the test failed it, the school came up short.
On the other hand, Arlington County's Long Branch Elementary School made it -- despite the poor performance of the 24 special-education students who took the Virginia tests.
The reason for the schools' unequal treatment under the law is that Virginia doesn't count the scores of any group with fewer than 50 students taking the tests. Maryland counts the scores of any group with at least five students. And that is only the beginning of the differences in their calculations, which even include different definitions of a passing test score.
The ambitious new law allows each state to set its own standards and devise its own tests to measure students' progress in math and reading, as long it makes "adequate yearly progress" toward "full proficiency" for all students by 2014. Schools that miss the mark two years in a row must develop an improvement plan or have one imposed. At low-income schools that receive federal funds, sanctions begin with allowing students to transfer and can end with a takeover by the state.
But as Deep Run and Long Branch illustrate, differing state standards make it hard to know exactly what a school has achieved -- and impossible to compare schools -- without reading the fine print.
About 36 percent of Maryland schools failed to make adequate progress last year under the law. So did 45 percent of Virginia's and 90 percent of Florida's schools but only 16 percent of Connecticut's and, according to preliminary estimates, about 20 percent in the District. That tells educators very little about the comparative merits of those states' schools.
"We can't draw any national conclusions from a national law," said Terri Schwartzbeck, a policy analyst for the American Association of School Administrators. When states do well, he added, "it could mean their schools are really good, or it could mean their plan made it much easier for a school" to succeed.
Critics of the law say this will create a race to the bottom. "Rather than helping to raise standards, it's going to lower standards," said Del. James H. Dillard II (R-Fairfax), chairman of the Virginia House Education Committee.
But supporters say the law builds necessary flexibility into its mandates to improve the nation's schools. No Child Left Behind contains the most detailed federal guidelines ever issued for education while respecting the states' traditional power to decide what goes on inside classrooms, said Eugene W. Hickok, acting deputy secretary of the U.S. Department of Education.
On the road to proficiency for all, the law requires states to track the scores of separate groups of students, including major racial groups, special-education students and those who are from low-income families or speak limited English. Each group -- and each school as a whole -- must hit state-determined pass rates each year, and those targets rise gradually until the deadline.
But states choose their own curriculums. States choose their own tests. In most cases, states determine the passing score on each test. States set the target percentage of students who must pass the tests to constitute "adequate progress." And states decide which racial groups are "major."
The students at Arlington's Long Branch took the Virginia Standards of Learning exams -- all multiple-choice at the elementary school level -- which have been in place since 1998 and are considered among the toughest in the nation. On the reading exam, at least 61 percent of Long Branch students, and 61 percent of each subgroup in each grade, had to pass to meet the standard for adequate progress.
But at Deep Run, students took the brand-new Maryland School Assessment tests. The reading portion included multiple-choice and short-answer essay questions. Only 43.4 percent of a school's students and subgroups had to pass the reading test to meet the standard for progress. (Like about half the states, Maryland uses a mathematical formula to move the target depending on the size of the group. The smaller the group, the lower the target.)
The District, which has not released its No Child Left Behind results, uses another system. Students take the off-the-shelf, multiple-choice Stanford 9 exams written by a national publishing company. There, 30.3 percent of an elementary school's students, and student groups, must pass the reading exam to meet this year's target for progress.
Each state's targets -- which vary by subject and sometimes by grade -- are based on a federal formula that takes into account how students have scored in past years. That helps explain why school officials in the District, where scores have been low, anticipate that nearly 80 percent of their schools will hit the modest target for "adequate progress" this year.
That also helps explain why only 55 percent of schools in Virginia, where scores have been improving steadily in five years of testing, made enough additional progress this year to hit the higher target. As Dillard put it, "If you're a state that sets very high standards, your chances of making adequate yearly progress on your own system are much lower."
To make comparison even harder, Virginia counts a subgroup's scores in the measure of a school's progress only if the group has at least 50 students tested. In Maryland, scores count as soon as there are five students in a subgroup. In the District, the number is 40.
And then there is the question of which groups to track. Virginia tracks only those that make up at least 5 percent of the state's population, so it does not count Asians and Native Americans as distinct groups when measuring progress. Maryland and the District count both.
The inconsistencies have frustrated educators, who have spent an exhausting year trying to decipher the law and figure out how to comply with it. "We're playing by 50 sets of rules across the country, but the penalties are the same," said Edgar B. Hatrick III, superintendent of Loudoun County schools and president-elect of the Virginia Association of School Superintendents.
Still, some said the law gave states less choice than it seemed. Virginia was forced to eliminate its exemption for immigrant children taking its tests for the first time. "It didn't feel that flexible," state Superintendent Jo Lynne DeMary said.
In Maryland, which spent $53 million creating its tests with an eye to the law, state Superintendent Nancy S. Grasmick said the law's strength is that it will force states, one way or another, to close the achievement gap between white students and minorities. "We have the children we have," she said. "They are the children we need to teach and help succeed."
The only alternatives to the inconsistencies, said Andrew J. Rotherham, director of education policy for the Progressive Policy Institute, are to "say we're just not going to worry about disadvantaged groups" or to create a single, national test.
"I don't think many people want that, either," he said.
Rosalind S. Helderman and Ylan Q. Mui
Comparing Schools' Progress Difficult
INDEX OF NCLB OUTRAGES