An op-ed in the Amarillo Globe News misses the point (annoying free reg required):
Tests don't test how much you really know but rather how much you know about what's on that particular test on the day you take it.
How this could be construed as a critical comment is a mystery to me. Does the author have some other option in mind that, perhaps, measures how much students will know in the future? Is there some magical alternative to tests that shows how much students will know next week? And is there any kind of assessment in the world that measures something other than what's actually on the assessment? Given that the asssessment can be made as broad or narrow as desirable, it seems odd to use this as a criticism.
Teachers must teach to tests or risk lower scores, which will ultimately affect the school's standing and the teacher's job.
Let me get this straight. Teachers must teach their students the material that their employers, the school/district/state, has deemed important, and students must demonstrate that teachers have done their jobs by performing well on exams. Otherwise, the state can conclude that the teachers aren't doing their jobs. Does anyone see a problem with this? Does anyone see why the role of "teacher" should be defined in such a way as to be mutually exclusive of the tasks above?
Some percentage of the student population can express what they know more fully using media other than standard tests; e.g., written essay, oral defense or demonstration.
Of course, some of those methods are so unreliable and difficult to grade properly that one could argue that, unlike with most tests composed of multiple-choice items, some students gain higher scores through these methods purely due to measurement error.
Oddly, the article then moves from the "criticisms" above to the perfectly reasonable argument that testing followed by no corrections, or by doubling inefficient efforts, are not useful. I just find it difficult to understand why the author spends several sentences bashing tests, and then moves to:
...what companies don't talk about, and probably couldn't even if someone asked, is how they use the results of tests to improve their operations. This qualitative distinction separates the successful from the unsuccessful. Those who use test results to highlight the need for action tend to do better than those who use them to punish.
But I thought tests don't actually measure how much you know. If the tests are that flawed, why do we expect that schools who use them to "highlight the need for action" will do better than schools who don't? Was the test-bashing in the first part of the article obligatory for someone seeking his anti-NCLB credentials? Or, like many test-bashers, is this critic merely trying to redefine tests (a la his call to abolish standardized tests in his conclusion) so that schools can use them only for feedback on individual students, and not to do those unhappy comparisons from school to school, or state to state?
Posted by kswygert at May 10, 2005 10:04 AM