Sample Opinion Piece:
A Tyranny of Standardized Tests
May 28, 2000
An Op-Ed piece from the New York Times
By Leon Botstein, President of Bard College
ANNANDALE-ON-HUDSON, N.Y. -- The good news about education has become obvious: the quality of public schools is now on center stage in national politics. From George Bush and Al Gore down to aspirants for state and local office, all politicians have embraced the cause of high standards and excellence. The bad news is that the remedy everyone but teachers and pupils wants to prescribe is more testing.
The mistrust of schools and teachers has become so widespread that the only politically viable solution seems to be to impose more standardized tests. Forty-nine states have now adopted curriculum standards that explicitly require testing in order to measure the performance of pupils, teachers, principals, and superintendents; by 2003, 26 states will have mandatory statewide tests for high school graduation. As a nation, we now administer at least 500,000 different kinds of standardized tests a year.
Testing has also become big business. We spend more than $200 million a year on it. The market for standardized tests is growing faster than that for textbooks. Unfortunately, the two go together all too neatly; three publishing companies dominate the market for tests and textbooks alike.
The tyranny of testing has become so intense that teachers may find themselves spending more than half the year teaching specifically for tests; their jobs and the standing of their schools are on the line. Not surprisingly, some teachers in New York and Maryland have been accused of cheating to improve their pupils' test scores.
The problem is that our mode of testing is primitive and out of date. We still adhere to the so-called "objective," machine-readable examinations pioneered in the mid-1950's. In the case of the Regents examination in New York, the questions themselves are often confusing or deliberately obscure and sometimes even embody errors.
Who, after all, is writing our English and mathematics tests? Not our leading writers, scholars and mathematicians. Furthermore, we still confuse speed with competence. Knowledge and understanding are not about rapid reflexes; learning is not a sport. Quickness of recall does not indicate depth of understanding.
Nor do all pupils know the right answer in the same manner or get the wrong answer or fail to grasp something the same way. Knowledge and skills are always approximations. A pupil who confuses World War I with World War II knows something more than one who mistakes World War I for the American Revolution. Today's testing instruments do not effectively account for how and what pupils know.
Even worse, they are not designed to measure the rate of change in each test taker. When we go for a medical checkup, we are evaluated not only in terms of an objective standard of health but on the progress or deterioration in our own particular bodies since the last examination.
The most egregious aspect of our mania for testing is that pupils never find out what they got wrong and why they got it wrong. High school students taking the Regents exams in New York do not get their tests back; neither do the vast majority of millions of children taking Iowa and Stanford tests or statewide reading exams. Most often, even the teachers don't get the results back in time to help them in their teaching.
What is the use of test results that are released months later, measuring a classroom, a school, a district or a state in terms of aggregate test scores? Would we tolerate a system in sports where the calls of umpires and linesmen remained a secret until the next season and the hits and errors of particular players were never revealed or justified?
In Texas, studies have shown that weeks after taking the Texas Assessment of Academic skills test created in the 1990's, pupils fail to show the apparent mastery of knowledge registered when they first took the tests.
Testing can and must be linked to learning. A mistake (and right answers) must be analyzed and corrected immediately for each pupil individually, just the way we respond when we teach sports and music. As it stands today, testing is little more than an adult political obsession that just results in more tests and profits for test makers.
Yet we now possess the means to change testing fundamentally. Rapid advances in computers and declining costs make powerful new technologies, once reserved for government and industry, accessible: technology involving complex computer simulation, as in pilot training, and manipulation of data. We can design tests that are interactive in a way that both helps learning and raises the standards of education. Even for young children, it would be possible to throw away the No. 2 pencils and machine-scanned answer sheets, and have the student tested at the computer itself.
There needs to be an initiative between government and the software industry to develop a new generation of tests. Programs can be written that inform the test taker immediately why the answer was right or wrong and that lead the pupil through the logic of the question to confirm understanding or correct ignorance. Politicians and school boards can still get their treasured measurements of timed test scores; as players are timed in a chess game, the clock for each test taker can stop and start as the individual goes through every question, discovering how and why he or she arrived at answers.
Computers also make it possible to measure the rate of change for each pupil -- indicating not only the student's progress but the teacher's effectiveness. Diagnostic tests at the beginning of each year can be designed that reveal what a pupil can and cannot do or does and does not know, establishing a baseline. If in the same classroom one fifth grader reads only at a second-grade level and another on an eighth-grade level, a teacher should be evaluated by what is learned by each of them over the course of a year.
Connecting testing to learning could also free teachers from forced adherence to bland state-approved textbooks. They would be given the opportunity to select and choose materials that meet the needs of each pupil. They would be able to justify the sort of differentiated, case-by-case decisions for which true professionals should be trained.
Reforming testing practices to make all this happen, however, will not be easy. Because software companies and the entrenched testing and textbook industry respond to short-term profit motives, public investment will be required for the longer-term work of adapting computer technologies to a new kind of testing.
Without a radical reform in the way we test, there will be no improvement in learning and educational standards. We can bridge the gap between those who are passionate about measuring standards and those who are ideologically opposed to testing as discriminatory and unfair.
What our politicians are offering us now is more of the same: a misguided reliance on a monopoly in a pseudoscience of testing that defines teaching and depresses learning.