“What was educationally significant and hard to measure has been replaced by what is educationally insignificant and easy to measure. So now we measure how well we taught what isn’t worth learning.”The truth of that isn’t acknowledged by Jeb Bush, Bill Gates, Lou Gerstner, Arne Duncan and other business leaders and politicians responsible for initiating and perpetuating the standardized, high-stakes testing craziness. They either can’t see or won’t admit the shallowness of their claim that “if you can’t measure it, you can’t manage it.” Challenged, they dismiss those who disagree with them as defenders of the status quo.
Using the scores on standardized tests to shape the life chances of kids, determine the pay and reputations of teachers, gauge the quality of school administrators, establish the worth of neighborhood schools, or as an excuse to hand public schools over to private, profit-taking corporations is, at the very least, irresponsible. If, as it appears, it’s a sneaky scheme to privatize America’s public schools without broad public dialogue, it’s unethical.
Figuring out how to measure original thought isn’t the only challenge test manufacturers need to address. Their tests:
- Provide minimal to no useful feedback to classroom teachers
- Are keyed to a deeply flawed curriculum adopted in 1893
- Lead to neglect of physical conditioning, music, art, and other, non-verbal ways of learning
- Unfairly advantage those who can afford test prep
- Hide problems created by margin-of-error computations in scoring
- Penalize test-takers who think in non-standard ways (which the young frequently do)
- Radically limit teacher ability to adapt to learner differences
- Give control of the curriculum to test manufacturers
- Encourage use of threats, bribes, and other extrinsic motivators
- Use arbitrary, subjectively set pass-fail cut scores
- Produce scores which can be (and sometimes are) manipulated for political purposes
- Assume that what the young will need to know in the future is already known
- Emphasize minimum achievement to the neglect of maximum performance
1 comment:
Exactly what I've been saying every time you've brought this topic up, Sam: machine-scorable tests can't capture anything but the least valuable products of learning.
To evaluate the important stuff requires simulation and extensive personal interaction with a subject-matter expert.
Only an expert can recognise expertise.
Post a Comment