Top Ten Reasons to Minimize (or Even Get Rid of) Standardized Testing

10.Standardized tests are said to be “objective” but are not.

Even on a multiple-choice test that is scored by a machine, the decision of what to include on a test, how to word test items and possible answers, and which answer is correct, is very much subjective. Some tests or test sections are not machine-scored fill-in-the-bubble; the scores on these are not only subjective, they are scored by poorly paid, under-trained temporary and part-time workers who are under major time pressure to score items quickly. Read about one scorer's experiences here.

9. Minorities and females under-perform.

According to Stanford University's “Stereotype Threat” study (among others), females and minority students perform worse than their actual achievement or ability would warrant, when compared to white males. (White males perform less well on math tests when they are told that their scores will be compared to Asian males' scores.) This under-performance is not because the tests are biased (although they are)—but because of anxiety caused by the stereotype that these groups perform at a lower level. Read more here.

8. Students from lower socio-economic groups under-perform.

No matter how carefully a test is constructed, it is going to have biases. Children with a lot of experiences outside of school and a lot of attention from parents score higher because they bring so much more previous learning to the classroom and the test. And kids with a lot of experiences and involved parents tend to be from middle-class and upper-class families—for the simple reason that adults in such families have more money to spend on their kids—art, music, and dance classes; travel; sports—plus more time and energy to spend with their kids. To find out more, read here.

7. Bad test items punish the most thoughtful and the most creative thinkers.

It is really hard to make multiple choice questions that are challenging, fair, and have only one unambiguous answer. Quite a few test items, therefore, have more than one answer that could be considered correct—especially for those students who “think outside of the box.” For more on this, read here.

6. Standardized tests have little predictive power.

Many colleges and universities give SAT results and other test scores very little attention, because there is only a very small correlation between higher scores and success in college. Some colleges and universities see so little value in test scores that they do not require testing, and almost all colleges and universities drop their test requirements for transfer students. Read more here.

5. High-stakes tests encourage cheating—even from teachers and administrators.

Many standardized tests are high-stakes. With school funding and ranking on the line, administrators tend to put pressure on teachers so that they will put pressure on the students. In some districts, teacher salaries and even their jobs get “left behind” if test scores don't go up. With money concerns tied up with the tests, there is temptation to cheat. Read more here.

4. Standardized testing takes up too much time out of schools' instructional year and out of kids' educations and lives.

It used to be that there was very little standardized testing, and most of it was of older students. Even today, many countries do little testing at lower grade levels. However, in the U.S., standardized testing starts early and is frequent, with some districts testing even 6 year olds, and some yearly tests taking up to two weeks at a time—and that's just for the administration of the test! (Test prep takes more time.) Even SAT tests take way more time, these days, than they used to, because it is common for students to sign up for test-prep courses and to take the SAT test several times. Mind you, all the hours spent taking tests could be used in other ways. Read more here.

3. High-stakes testing causes stress and anxiety for all involved.

In #5 above, we mentioned that tests can be high stakes (think money) for schools, administrators, and teachers, but what about students and their parents? In addition to the pressure put on students by schools and teachers, they often face high stakes like graduation, college acceptance, college credits, and scholarships (which can be translated into money, money, money, and money). With high stakes, test anxiety is natural—but anxiety not only causes lower scores (see #9 above) but also gets in the way of true learning. Read more about negative emotional effects of testing here.

2. Standardized tests—particularly those that use multiple-choice format—don't measure the most important facts, concepts, or abilities.

Because test scoring mandates that students are differentiated from one another, the test items do not cover just the most important facts that most teachers focus on—because “too many” students get those items correct. Therefore, tests include many less-important facts. Unfortunately, some teachers and schools, in an effort to boost test scores, then start to teach those facts as if they were important. This teaching to the test can water down curricula in unintended ways.

Also, standardized tests are almost useless in measuring true understanding, positive or negative attitudes about a subject, problem-solving abilities, and upper-level thinking skills. Because test scores have become very important to schools, teachers, parents, and kids, true understanding, problem-solving, and high-level thinking have all been reduced in importance in education.

Read more about this topic here, or in the PDF file “The Case Against Standardized Testing,” by Alfie Kohn. (Linking doesn't work for a PDF file, but you can Google it.)

Finally, here is the Number 1 reason to limit or eliminate standardized testing:

1. Low test scores rarely lead to improvements, and the very structure of testing ensures lots of low scores.

One rationale for giving tests in the first place is to hold schools accountable for educating children, and to show teachers and parents weak areas that need improvement in individual children. Given that reasoning, we would expect to see the feedback of low scores in certain academic areas motivating the school to experiment with new instructional methods and the government or school district to provide extra help to the school in that area. In the case of just one particular student, we would expect to see remediation, again by changing the learning situation, trying new modalities and techniques, and by providing additional time, effort, or personnel to address the deficit.

However, low scores really just become an excuse to punish. Schools with low scores tend to lose resources rather than gain support. Students are often put into “lower” classes with less accomplished students and (often) teachers, thus perpetuating the lack of success.

We do not see feedback from standardized tests used for improvements in schooling, teaching, or learning because, it turns out: Tests are just a way of ranking students and schools.

The very design of these tests as norm-referenced instruments requires that half of the students taking the test are judged “below average”—no matter how accomplished they may be! The need to appear useful, to seem to be measuring something worth knowing, pushes test makers to deliberately make tests that will create a wide range of scores, and of course these scores are arranged on a bell curve of success at one end, failure at the other, and mediocrity a large bulge in the middle. Read here to find out more.

What good does it do to find out where children are ranked? There aren't many good results from ranking kids, but there are plenty of bad results for individuals and therefore society: Anxiety. Low self-esteem. Lack of motivation.

Some of the worst problems come from the simple fact that testing begins so early—when many of the students being tested are not ready to do the tasks on the test. The age of reading readiness varies, and the ability to conceptualize number does not necessarily coincide with readiness for paper-and-pencil arithmetic. Young children who aren't developmentally ready for formal academics are pushed into a downward spiral of negative expectations by early and frequent evaluation.

As a thought experiment, imagine kids facing “standardized tests” about walking and speech when they are just babies and toddlers. How would it be helpful to rank the babies and infants according to their accomplishments at age 9 months, and 11 months, and 13 months, say, for walking, and 16 months, 20 months, and 24 months, for talking? Would that help parents and babies or only lead to more stress and quite possibly walking and talking “disorders”?

***

Standardized testing has grown to be big business! It has taken on a life of its own in the minds of politicians and journalists and society at large—but it is based on an outmoded theory of learning rather than the latest findings in child development, psychology, cognition, and education.

Standardized testing as it exists today does more harm than good, and it should be greatly limited. Or even, quite possibly, abolished.