Dayton Daily News education reporter Scott Elliott writes about schools, kids, teaching and learning.
Home > Blogs > Get on the Bus > Archives > 2006 > July > 14
Friday, July 14, 2006
Trusting tests over teachers
Michael Winerip, the education columnist at the New York Times, created a bit of a stir in the blogopshere this week when he wrote his final column.
Winerip is a great journalist and a wonderful writer. But he has irked the pro-standards crowd with criticism of No Child Left Behind. His last column is a parting shot at NCLB in which his chief complaint is that the law’s primary message to teachers is that we don’t trust them.
Winerip argues that an education reform can only be successful if it has buy-in from the people that will put it into place. There are lots of other dimensions to the piece. For instance he trumpets small classes as a key solution to the problems of education and needles lawmakers by suggesting they be evaluated in the same way NCLB evaluates educators:
“We need a No Family Left Behind Law. This would measure economic growth of families and punish politicians in charge of states with poor economic growth for minority families.
FOR example, in Ohio, black families earn only 62 percent of white household income, one of the biggest disparities nationally. So every year, under No Family Left Behind, Ohio would be expected to close that income gap. If it failed to make adequate yearly progress for black families’ wealth, the governor and legislators would be judged failing, and after five years, could be removed from office. This way public schools wouldn’t be the only institutions singled out for failing poor children.”
But back to the question of trust. Here’s what he says about the way NCLB relies on tests instead of teachers:
“Because teachers’ judgment and standards are supposedly not reliable, the law substitutes a battery of state tests that are supposed to tell the real truth about children’s academic progress.”
There are convincing arguments on both sides of the question of standardized testing. But Winerip hits on a very narrow question that I find interesting — why do we trust the tests?
Whether a child gets a low or high score on a standardized test, what does that mean to us? That the child is smart or dumb? That they are well educated or poorly educated? Those are the sorts of judgments we often make based standardized test results.
But are those the questions the tests were designed to answer? Do the test makers claim that these tests will give us those answers? (I’ve asked them. The answer to both questions is no.)
And what do we really know about standardized tests and how they are made? Can we be sure certain they are fair and that they accurately measure the material they claim to test?
We’ve written a good bit about testing here at the Dayton Daily News. What we’ve reported is that the methods for creating, scoring and setting passing scores for standardized tests are flawed.
There are too many bad teachers out there, that’s for certain. For some, standardized tests are a solution to that problem. But the flip side is that there are too many bad standardized tests out there, too.
It is an interesting and debatable question, then — how do when know when to trust a standardized test instead of a teacher? And vice versa?
That tough question, I think, is under-reported and under-discussed. I’d love to hear your answer.
Permalink | Comments (4) | Categories: Testing
