Erasures demonstrate huge sensitivity in ratings

The Dispatch had another speculative piece of reporting on the attendance erasure issue. We'll leave educator Greg Mild at Plunderbund to go over the substance, or lack thereof, of the article itself. We want to concentrate on something else mentioned in the article which stood out.

At the heart of the controversy is this

Though 7 percent [number of deleted records] may not sound like a lot, it could have a big effect: There are students behind those numbers, and some of their standardized test scores were likely discounted when their attendance records were deleted. That means Columbus’ school grades could have been artificially inflated because of records-tampering.

From this, the Department of Education had a remarkable comment

“The math indicates that removing one student could affect the overall rating,” said John Charlton, spokesman for the Ohio Department of Education. “In some districts, it could be one kid. It’s about dropping the right kid.”

Can this be true? Can the ratings of an entire district truly be affected by just a handful of students, or even one? We have an actual true life example, that yes, small numbers of students can indeed cause an entire districts rating to be changed

In Lockland near Cincinnati, removing 36 kids from the rolls — about 5 percent of their student population — lifted the district from a C to a B on the state report card.

Can a school rating system that is so sensitive to just a handful of students truly be measuring the district as a whole? This growing controversy over attendance data is revealing a lot more than people realize.