standardized test

A test where the answers are standard (multiple choice, true/false, etc.) and can be graded by just comparing the answers with a key. Usually taken by filling in bubbles on answer sheets so that the test can be graded by a computer.

Not usually as good a measure of the student's knowledge than an essay or project where they have a chance to explain what they've learned, or why their answer is different from the other students'. This is especially true when the test is being given to very large groups containing different schools, cultures, or native languages.

However, a small number of standardized tests such as the AP (Advanced Placement) exams do have a component that requires you to write down your thoughts or the process you used to reach an answer, but these aren't terribly common as they are time-consuming and labor-intensive to grade.

Any one of a number of tests that measure your ability to select the correct answer from a series of options multiple choice true/false matching. Your results are then compared to a sample population and you are told how you compare.

There is a growing movement in the educational community towards free response or essay questions. There is a belief that these evaluations are a better measure of competance.

However, since essay tests are difficult to evaluate and time consuming when administered nationally, the ScanTron appears to be a permanent resident of American education.

<$0.02>

Standardized Tests should be banned from public school in the United States (I cannot speak for other countries, I am not a product of their respective school systems). I'm not arguing whether or not they're an accurate measure of learning, or their effectiveness. The thing that makes me angry, is that the makers of the test are deciding what I learn.

My math teacher has skipped entire chapters of the textbook, areas that aren't covered on the standardized test. What I am allowed to learn is being controlled, what I am allowed to study is being bordered.

It gets worse, much worse. The test makers begin to control what I learn, they standardize it. This reflects upon the teachers, forcing them to conform, and teach the test. When teaching becomes standardized, when everyone is taught the exact same way, when we lose that precious diversity, thoughts are unborn, ideas are unborn, and we cease to learn from each other.

Education should teach kids how to think, not what to think.

</$0.02>

All of the above writeups, although expressing the appropriate "screw the system" ideals, fail to mention the major aspect of the standardized test: they are standardized.

Let me elaborate on what that means. These tests came about as an attempt to figure out whether a student had learned a certain body of material. Since all elementary schools, middle schools, and high schools are different (and some children are home-schooled), various educators have attempted to create a device to figure out whether you've learned, say, algebra, regardless of where you were supposed to have learned it.

Put yourself in the shoes of a college admissions person. Let's simplify this scenario by saying that you're only concerned with whether the students you admit have learned the following topics:

• basic algebra
• basic trigonometry
• basic geometry
• basic arithmetic
• certain concepts that lead up to calculus, such as functions, inverses, conic sections, the concept of rates of change, etc

Easy, right? Make a big exam that asks questions from a selection of topics. Algebra is algebra. Either you know about it or you don't. Same with the other topics. Standardization just means that the test has broad applications, and that a student's score will be statistically analyzed to show how well they did with respect to all other students that also took the test. This is intended to eliminate bias incurred from problems with the test, as well as external factors such as lack of time, broken pencils, faulty calculator batteries, narcolepsy, etc. The statistical methods involved are quite sound (if you don't think so, get a good book on statistics). As with all statistics, however, their meaning is often misinterpreted.

In my home state of Arizona, for instance, there is a major dispute ongoing concerning only granting a high school diploma to a student if he or she passes a test. Because public education in the US is on a state-by-state basis, and they want the diplomas of all Arizona public high schools to mean the same thing, they make the test applicable to all schools.

The results from the first round of AIMS (the name of the test) were abysmal. Around 25% of students from most schools tested passed the math section! But rather than question, gee, why aren't these students being taught how to multiply fractions, parents kicked up a huge fuss about how the test was unfair, because surely their little angel would never fail a test. Other children -- not their children -- are the ones that don't know how to multiply fractions. Their kid is just as smart as all the other kids.

The point here is that enth is right. Your SAT score only measures how good you are at taking the SAT. However, the math section of the SAT is intended to measure how comprehensive your knowledge of basic mathematics is. "Standardized" tests are no more evil than any other exam you might have in a class, the only difference is that they are used to generate statistics, and thus are not tailored to your particular educational background.

Don't get me wrong. I have strong objections to the "verbal" section of the SAT and similar tests. You can't test someone's communication skills based on that kind of rubbish. The AP English examinations are more on the right track, but these will cost you around US$70 a pop, reflecting the labor-intensive grading process and the market for reliable analysis of prospective college students. But I fail to see the problem with using standardized tests as a criterion for determining mathematical competence, because, contrary to the beliefs of certain liberally-inclined pseudoeducators, either you get the right answer or you don't. If you get 9 of the 10 equation-solving problems wrong, we can conclude that you don't really know what you're doing. This is sound logic.

Schools "teaching the test"? Great! That means that they might learn how to multiply fractions. Does this confine your learning boundaries? Yes, it does. It means you might have to actually learn some math instead of "discovering" some properties of cardboard cones, which is what starts happening when the "Everyone's A Winner" mentality starts encroaching on education. Heaven forbid someone should tell you that if x + 10 = 10, then x = 0!

And actually, I think public education is a bad idea any way you set it up. But in the meantime, keeping tabs on the effectiveness (assuming there is any) of certain curricula is the only way to provide any sort of quality assurance.

State-mandated Standardized tests, when used as a primary measure of student success, are as damaging to schools as they are to students. A student can run into problems due to poor test grades, but he or she can recover in college. However, a school district marked as failing may never recover.

This is not due to any inherent flaw in the tests themselves; it is due to the way the tests are used. Instead of using the tests to judge where students are and what needs improvement, tests are often used to judge the success or failure of a district. However, the tests are wholely inadequate for this task, because of the number of variables at play. Lets look at an example:

In my home state of Ohio, the state Proficiency Tests were used until very recently. Students took these tests every few years in Math, English, Science and Social Studies. Students who failed could be prevented form graduation whatever school they were in (elementary, middle or high school). However, the school district was also graded, and depending on how many sections sufficeint numbers of students passed, the district would be rated. These ratings could be used to determine funding and the level of autonomy districts were allowed by the state - if they did poorly, the state took over the district. Unfortunately, no effort was made to correct the scores for the circumstances surrounding the schools. This resulted in an obvious problem; regardless of the quality of the teachers, schools and resources available, schools in richer areas scored better and schools in poor areas did worse. In addition, districts with largeminority populations tended to do worse, because in Ohio minorities make are disproportionately poor.

My district had a lot going for it - nationally renowned fine arts programs, a well-developed and effective program for gifted students and many well-educated and a few nationally recognized teachers. Did the district do well? No. We had the geographic disadvantage of bordering Cleveland.

Our district was so good that students from the innercity and other, poorer inner-ring suburbs to go to our schools. Every year the Central Office would discover dozens of students fraudulently attending the school. This dragged down our scores.

Regardless of this disadvantage, we still did well. Our Caucasian students and minority students both scored better than the state average and comparably to the top districts in the state. However, we still did not score well and, as a result, placed in the category of "academic watch." Parents and community members saw this and did not bother to ask many questions - if the state said so, it must be true. As a result, the district finds it difficult now to pass school levies because parents are not satisfied that it is doing a good job, despite the fact that closer examination of the scores would reveal that we are doing quite well.

The biggest problem that arose was the tendency to use the test as the sole measure of the district when it could not possibly do so effectively. Expecting a school district to instantly raise any student transfering into the district to state standards, regardless of that student's previous performance, is foolish. However, the grace periods allowed schools to bring students up to speed are ridiculously short. In addition, the effects of the district's population is ignored, despite the fact that it may well be more impressive for an inner-city school to teach all of its student to read at grade level than it would be for a wealty private school to have all of its students score 1600s on their SATs.

Standardized tests have their place - for example, they can be used to get a rough estimate of a student's ability and can help identify problem areas. However, they are rarely used in this way.

What a standardized test cannot do is serve as the sole, final judge of anything. Standardized tests can produce skewed results, be misleading or be flat-out wrong. However, they are easier than doing effective assessments, so they are used.

In their 2005 book Freakonomics, economist Steven Levitt and journalist Stephen Dubner begin their self-defined exploration into “the hidden side of everything” by examining what comparisons can be made between public school teachers and sumo wrestlers, comparing the expectations for honesty and respect each profession garners. Then they examine one other disturbing similarity found between the two groups, one that happens when sumo wrestlers are one win away from moving up in their rankings and when teachers are administering standardized tests. Both of these situations are crucial moments: a poor outcome could affect the professional’s career for years, if not end it altogether. And so Levitt and Dubner find evidence of one other, worrying similarity:

In situations such as these, where the incentives are strong enough, both public school teachers and sumo wrestlers will cheat.

In the case of the teachers, this cheating can either be rather direct - putting the answers to the standardized test on the board, changing student’s answers in the brief amount of time allowed before the tests are handed in - or indirect, if the teacher provides more assistance than permitted by the testing agency.

This is awful, but understandable. The teacher can be subject to severe penalties if their class fails to pass a certain grade on the test, or even if it fails to improve by a certain amount from last year. In some districts, teachers’ bonuses are directly tied to their classes’ performances on these tests. And this obsession with the test comes from the negative side effects of a poor performance on the test - the entire school district could be audited or denied resources if they’re found to be underperforming.

The problem is, the motivation for standardized testing is pretty easy to understand. Grading can be a very subjective process, and it can be influenced by a multitude of external factors. This subjective process is then performed by a variety of teachers in various schools in various districts - each with their own idea of what a “correct” answer is, what the various grades mean, and how things such as a GPA should be calculated - which each have their own incentive to try and inflate these grades as much as possible. Outside of the closed loop of a single school - and, possibly, a single class - grades make for a meaningless unit to be compared. Tests provide a baseline number, an easily compared unit which can be used to make various judgements. And so long as you ignore how arbitrary and fleeting that number is, you can justify any sort of policy with it.

The movement against standardized testing is strong and growing, uniting parents and teachers under a common banner. I’m with them. Standardized tests are awful, penalizing teachers and districts unfairly and giving an advantage to only one sort of student. The problem is, calling for a system’s downfall is easy. Proposing a viable alternative is much, much harder.