The Myths of Standardized Tests: Why They Don't Tell You What You Think They Do - Page 3 of 4

The Myths of Standardized Tests: Why They Don’t Tell You What You Think They Do

The current NCLB system and Obama’s Blueprint place excessive reliance on rewards and punishments. The assumption is that they will motivate students and educators. It relies on a belief in behaviorism, which focuses only on objectively observable behaviors and discounts mental activities. They do not take into consideration the challenges that confront individual schools. Sanctions do not tell a staff what is wrong or how to improve the situation. There is no evidence that sanctions are effective. There is also research that shows how rewards dampen interest in, and enthusiasm for, the activity that is rewarded. Fortunately we haven’t been dangling many rewards in front of kids. Human motivation is far more complex.

Campbell’s law states: “The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures, and the more apt it will be to distort and corrupt the social processes it was intended to monitor.” So if the stakes are high, people will work harder to improve. Such hard work can subvert the accuracy of the measurement. If test preparation is excessive, the tested students will no longer compare to the group on which the test was normed. Schools caught cheating are only the tip of the iceberg. Some schools improve scores by counseling students to dropout into GED programs. Educational triage takes place when teachers focus on “bubble kids” who will pass with help at the expense of others. Lots of test prep takes time from non tested subjects.

Although most educators think depth is more important than breadth of coverage, the proliferation of state-level standards results in the polar opposite. Teachers survey entire fields to “cover” the curriculum and stuff kids like turkeys with as many facts as possible. The impact is even starker in areas that are not tested. High-stakes tests have narrowed and dumbed down curricula and eliminated time spent on untested subjects by about a third, including recess. The fact that this happened was not part of discussions that lead to the tests in the first place. Tests also drive instruction towards items with one clear, right answer at the expense of open-ended problem solving. Explaining is hard to do when you are blackening a bubble. Common Core Standards being pushed by the Obama administration only contribute to the problem.

Standardized tests are indirect measures of knowledge and skill. The authors cite the road test that new drivers take as a direct measurement of the skill to be accessed. It is a test of performance in the real world built on knowledge judged by a written test. If a system featuring paper-and-pencil tests with performance tests is so good, why don’t schools use it more often? The first reason is the cost of direct measurements like the measurements required by special education laws. We forget that cost and value are not the same. The second is bureaucratic inertia, which simply accepts and builds on past practice. For example, colleges require GRE scores, a supposed predictor of grad school success, for students who already have graduate degrees. Standardized tests may be reliable, but they are not valid measures of performance.