Many black kids' tests left out of racial groups

About 40,000 pupils in Ohio aren't counted under racial categories.
COLUMBUS (AP) -- LaTonya Robinson knows her son did well on tests at his elementary school. What she didn't know was the limit the school put on reporting his results.
Federal law allows Clinton Elementary School to exclude the test scores of black children as a separate racial group -- including those of her son, James -- when measuring how well those children did on reading and math.
In Ohio, almost 40,000 pupils, or about 4 percent of all test scores, aren't counted under the law's racial categories, an Associated Press computer analysis found. Almost all were minority pupils, and most were black or Hispanic.
Overall, AP found that about 1.9 million pupils -- or about 1 in every 14 test scores -- aren't being counted under the law's racial categories. The scores of minorities were seven times as likely to be excluded as the scores of whites, the analysis showed.
At Clinton, the scores weren't counted last year, James' first year at the school, or in 2004, when only 27 black children took tests there -- below a statistical cutoff for using the scores to judge whether schools met federal requirements.
"I don't think that should have happened," said Robinson, a state government auditor. "Everybody should be counted."
A key element of the federal No Child Left Behind Act was tracking pupil performance by race to ensure that all children, black or white, Hispanic or Asian, were making progress.
The goal was overcoming what President Bush called the "soft bigotry of low expectations."
Under the 2002 law, all pupils, including subcategories broken down by race, must achieve proficiency in reading and math by 2014. All scores, including those of minority children, must be included in the overall measure.
Schools receiving federal aid also must demonstrate that pupils in all racial categories are meeting performance goals or risk penalties that include extending their school year, changing their curriculum or firing their administrators and teachers. If any category of pupils fails, the whole school fails.
But that requirement came with a loophole. Using a cutoff, schools don't have to break out the scores of racial groups that are deemed too small to be statistically significant.
In Ohio, the number is 30. It's much higher in other states.
-- 52 in Oklahoma -- and much lower in others, such as five in Maryland.
In Ohio, black children constituted 35 percent of the total number of schoolchildren whose scores were excluded from reporting by racial group, followed by Hispanic children at 29 percent.
A list of schools in Ohio with 29 black children tested in 2003-04, just under the cutoff, meaning they excluded those children's scores from racial categories, includes Mahoning Unlimited Classroom, Youngstown. The school is a virtual charter school.
Comparisons
By contrast, white children, the vast majority of Ohio schoolchildren, constituted just 10 percent of the total number of excluded children.
Put another way, 8 percent of black test scores in Ohio aren't broken out at the school level, while only 0.5 percent of white test scores aren't broken out.
Critics of the exclusion say schools can meet their federal mandate and no one would ever know how minority children did.
They say this hurts the state's ability to address the recognized gap between the scores of white and black children.
For example, the AP found that in Ohio, 81 percent of the 1,829 schools able to exclude the separate reporting of black children's scores met overall annual yearly progress goals -- the measure by which schools pass or fail under No Child Left Behind.
"If you're going to really look at the achievement gap, and you're going to look at servicing all kids, why if certain schools fall under a certain number are they allowed to negate those children?" said Deb Tully, director of professional issues for the Ohio Federation of Teachers, the state's second-largest teachers union.
Educators strongly disagree with this conclusion. They say that though the scores aren't reported as a subgroup, districts are aware of how each child is doing.