The Fourth Form music class from London's Islington Green School provided that spirited chorus to Pink Floyd's 1979 rebel anthem, Another Brick in the Wall. I doubt the song would sound so good today--at least if American fourth graders had to sing it--because our kids are too busy to sing. Instead, the children have tests to take and re-take. And study for. And watch the adults in their lives obsess over. Otherwise the State of Texas could not assess the kids' "academic readiness."
As some of you may know, Mrs. Nester is ending her first year as an assistant principal at a Waco ISD school. She serves an economically disadvantaged population with almost the entire school qualifying for the free and reduced lunch program. This position comes after long years teaching at some of the highest-achieving schools in the state. Through her eyes, I have now seen the high-stakes testing picture from both sides and it is not pretty.
Starting almost two decades ago, the Bush brothers popularized the idea of "Accountability in Public Education." Jeb brought his reforms to Florida; George W. spread the message to Texas and then across the nation. As a general principle, I applauded their efforts. Anyone spending tax dollars should be responsible for getting results. On top of that, a high school diploma should mean that the graduate is literate and owns a basic understanding of how our world works. Finally, President Bush's signature education accomplishment, the "No Child Left Behind Act" stands for the noble proposition that the location of a child's school should not determine how well the child is prepared for life, at least academically speaking. It is a list of tasty ingredients: accountability, competency and equality. Unfortunately, we have overcooked the dish to the functional equivalent of school cafeteria spaghetti, served cold.
The criticism of high-stakes testing has always been that the tail would wag the dog; schools would stop educating and become test preparation centers. That prediction has largely come to pass; the fact that school test results have become the be all and end all of the education system is close to self-evident. Perform poorly enough as an administrator or teacher and you find yourself looking for new employment. Fail the test as a student and you look down the barrel of another year in the same place. If you teach in Waco ISD, the goal for each student is a passing grade; if you teach in Highland Park ISD, the goal for each student is a "commended" mark. Either way, we judge our schools based almost exclusively on the test results. In that world, school administrators understandably direct their efforts to maximizing scores.
The answer to that criticism has been basically "Good." Meaning that, if we design the test right, good results mean students have learned what they should. Considerable drama accompanies the question of whether we can actually write tests that measure "what students need to know." For example, to obtain a high-school diploma a Texas student must pass five end-of-course exams, encompassing English I & II, Algebra I, Biology and U.S. History. In the legislature's first testing iteration, however, 15 tests served as the measure of "sufficient knowledge." Included among the 15 tests were English III, Algebra II, Chemistry and Physics. Responding to public outcry, the legislature pared the number of tests back to the five now required. Was the reduction in test requirements an example of common sense, recognizing that a high-school diploma should only mean that the holder can function in every day life? Or did the reduction say Texas was lowering its sights, retreating from the academic excellence necessary to drive the world forward? Both versions have merit and support. Whichever view one accepts, the truth is that the decision on what to test is political; decided on the state, rather than the district, level; and usually obscured from public scrutiny. And I have purposefully avoided the question of whether the left or right wants us to slant the tests to accommodate one agenda or another. So we start the process from a place of distrust: do we even know what we are trying to measure?
Let's assume, however, that we can agree on what to test. One would think that if we were going to base our entire system on test achievement, we would administer the test in a way that produces verifiable results. One would be wrong. In evidence-based inquiries, the people or entity running the inquiry must be objective. If this were a criminal or civil trial, the community (the "jury") would be given the scores (the "evidence") and render its verdict. Fair trials are run by good judges, who in addition to being knowledgable about the law, are independent and objective. In this trial, however, we have the schools administer the very test they are being judged on. In other words, there are inherent questions about any "good evidence" we get.
There are of course, numerous examples of overt cheating by administrators and teachers. There are more subtle ways to "game the results" for your school based on which students have their scores reported. It is common practice for schools at the top of the food chain to rigorously enforce attendance zones against at-risk children, yet look the other way when it comes to the smart ones. But the biggest downside to having school personnel administer the tests is that if a principal or teacher produces an outstanding result, our question is not "how can we replicate and broaden these results?" but "how do we know these improved scores are real?" Finally, given our desire to slice and dice the data in every imaginable way, "administering the test" involves much more than handing out a question booklet and an answer sheet. Mrs. Nester spent countless hours training to be a test administrator, inputting and checking test-taker data, and implementing a security plan. Actually by "countless" I mean "uncounted." I would hazard a guess, however, that if someone had counted her time, they would have found that Mrs. Nester spent well over 100 hours doing administrative work rather than helping children learn.
The next major issue is the question of what it takes to pass the test. The assumption underlying the whole system is that if children have the knowledge, they will answer the question correctly. That assumption is probably not a sound one. In elementary school, the test can take hours. Anybody who thinks that one can measure the attention span of a third-grader in hours does not spend much time with third-graders. As to the upper grades, I was recently reading Malcolm Gladwell's excellent collection of essays What the Dog Saw. In his story detailing why people sometimes do not perform to their potential, Gladwell makes the point that groups subject to a negative stereotype perform better on tasks within that stereotype if they feel they are not being judged. Yet the system makes it completely clear that people are judged through the tests. In other words, a host of psychological factors contribute tremendous "noise" to the correlation between a test score and actual knowledge. Anecdotally, any teacher can tell you about the bright student who just does not care about the test.
As an antidote to test anxiety, many schools have altered their basic teaching strategy. Rather than trying to deliver knowledge and critical-thinking skills to their students, teachers spend all day coaching their charges to beat the test. Children no longer read books and discuss or write about those books in class. Instead, students work endless passages designed to increase their comfort level on the test. Somehow we have decided that test-taking strategy is more important than a love of learning. And then we scratch our heads trying to figure out why kids will not behave. Really?
My criticisms are not new or novel. If you have reached this far, you might be wondering why I bring it up now? Because having watched a good person fight as hard as possible for a year, I have gained a personal understanding of the folly being perpetrated. Mrs. Nester's school year is ending except that she has 38 fifth-graders who did not give enough correct answers and will forfeit the better part of June in light of their failings, with Mrs. Nester there to guide them. We do not know how the third and fourth graders did. But we do know whether they passed or not, the majority of children are not engaged learners. In a 21st century replete with technology that makes the world a more interesting and accessible place, schools have become test mills that maximize tedium. Good-hearted teachers and administrators end the year beaten and dispirited, trying to understand why they cannot achieve the impossible.
So in the spirit of progress, I humbly offer the following solutions:
- No elementary school high-stakes testing
- Tests to be administered by a third-party or state agency
- Design school-evaluation criteria that captures student engagement and place those results on equal footing with test scores
- Buy a teacher a margarita this week.
I have no power over the first three, but I will make good on the fourth.