Commentary

Education Law Encourages Fuzzy Math

By Marie Gryphon
This article appeared in the Newark Star-Ledger on February 18, 2005.
Recently the state of North Carolina hired a consultant to learn whether its teachers and administrators are cheating on high-stakes tests required by the No Child Left Behind Act. North Carolina’s decision follows recent revelations of misconduct elsewhere, ranging from asking weak students to stay home on test day to erasing wrong answers and replacing them with right ones after the fact.

The Indianapolis Star recently reported the suspension of a teacher for pointing out wrong answers to her third grade class – the third such incident in Indiana. The Dallas Morning News published several investigative articles uncovering organized misconduct in dozens of Texas public schools and suspicious scoring patterns in hundreds more.

State and federal politicians quickly claimed a rare opportunity to seize the high ground. They’ve promised to stamp out small-time mendacity by marginal teachers in struggling schools. But they themselves are committed to a higher-stakes bluff: they must convince us that the No Child Left Behind Act can work.

The Act wasn’t designed to work. An accidental byproduct of bipartisan compromise, it reflects no single idea of accountability. Lauded as a great political achievement, it is unlikely to improve student achievement. And like an emperor’s nudity, no state or federal functionary can afford to notice the fact.

States must generate the appearance of complying with the law, including a display of “annual yearly progress” necessary to stave off the Act’s more discombobulating remedies. They are doing so by fudging their figures.

Some states play this game better than others. According to the RAND Corporation, Texas boasted an 88 percent pass rate on its eighth grade reading test last year while South Carolina turned in a miserable 21 percent pass rate.

Texas children read far better than South Carolinians, one might conclude. One would be wrong, though. On the standard National Assessment of Educational Progress, scores from these two states are nearly identical: South Carolina has a 24 percent “proficiency” rate compared with only 26 percent among Texans.

Different state exams were useful in a pre-NCLB world. Then, states set standards with an eye toward cleaning their own houses. Now they tailor tests and statistical methods to obscure reality, compromising the integrity of state systems to keep the federal government flying blind.

Last year the state of Michigan reduced the number of “failing” schools under its care from 1,500 to 216. But this remarkable achievement was merely a statistical sleight of hand. Michigan lowered the minimum passing score on the state’s assessment from 75 percent to a mere 42 percent, the Heartland Institute reports.

Other states lower standards by manipulating their methods for reporting results. North Carolina increased the percentage of its schools reporting adequate progress from 47 percent to 70 percent last year. The increase is largely due to a technical change in the way the data are reported. Kentucky, Pennsylvania, and Tennessee have made similar adjustments.

But state bureaucrats aren’t the only ones with reason to dissemble. The Bush administration touted NCLB as a seminal domestic policy achievement. It wants the Act to appear successful, and it seems prepared to tolerate fuzzy math.

The Department of Education has been complicit in these state shenanigans and others, approving technical obfuscations and plans that backload required progress to a time when both state and federal executive branches are sure to be in different hands. Just this week, the Department reversed a prior decision to agree with North Dakota that a teacher can be “highly qualified” by no other standard than having spent years in a public school classroom.

NCLB has states and feds smiling thinly across a mountain of paperwork like two cheating spouses in front of the children. Federal micromanagement of state testing regimens just wasn’t a very good idea.

The foregoing facts have inspired demands that Congress dump state tests in favor of national assessments, or empower federal bureaucrats with more minute control over schooling. But it is no solution to demand that the federal government toughen up. NCLB has already caused some homogenization of school curricula, a disturbing trend that may thwart fruitful experimentation by states and localities.

The No Child Left Behind Act should be ended, not mended. If the law were repealed, Congress could unearth some good proposals, like the House measure that would grant massive regulatory relief to “charter states” experimenting with real reform. Better yet, lawmakers could revisit an older idea: a plan to phase out the federal department entirely.

Marie Gryphon is a policy analyst with the Cato Institute’s Center for Educational Freedom.