Standardized vs. Standards-Based TestsLesson Plans > General > Assessment
Standardized vs. Standards-Based Tests
Much is being said these days about standards-based education and standards-based testing. Unfortunately, the terminology is not always well defined. Sometimes educational jargon can be used in different ways by different people, leading to a great deal of confusion.
Two of these confusing terms are "standardized test" and "standards-based test." What does each of these mean? This article is an attempt to clarify the terms.
We are all familiar with standardized tests. IQ tests, CAT 5 tests, and SAT tests are all examples of standardized tests. A standardized test is one in which a student's performance is measured in comparison to everyone else who took the test. If everyone did poorly on a standardized test, and you happened to be the highest of those low scores, you automatically score the highest possible score in the band. If, on the other hand, everyone did very well and you happened to be the lowest of those high scores, then you automatically score the lowest possible score in the band. This is true, even if you achieved the same level of performance each time. In other words, your rating is dependent on everyone else's performance.
Consider the phrase "grading on a curve." Most people assume that grading on a curve means scaling the test. In actuality, grading on the curve adjusts scores so they fit the classic "bell shaped" curve used in statistics. In this curve, the very upper end is usually assigned A's, the next section down is assigned B's, the large central section is assigned C's, and next lower section is assigned D's, and the very low end is assigned F's. The point is to insure that there is the same number of F's as A's, and the same number of D's as B's, and that most people score in the C range.
Suppose a teacher gives an exam, and then grades it on the curve. Furthermore, assume that the highest score on the exam was a 40%. The highest scoring student receives a grade of an "A" for the test, even though he or she scored an extremely low grade. "Well, that's okay. Obviously the test was much too hard, so it's reasonable to give the highest grade an A."
Now consider the reverse scenario. The tests are scored, and the lowest grade is an 80%. Because it is the lowest score, it receives an "F". "That's not fair!" is the cry. "Obviously everyone understood the material well, so you shouldn't punish this person".
In each situation, the same process is applied. One of these is considered to be acceptable, and the other is considered to be unfair. This is the fundamental problem with standardized testing. If properly used, standardization is applied to a very large group of people, and students see only their "percentile", not their "percentage". When used in this manner, standardized tests do have some value. However, the standardization process has far less value when used within a teacher's classroom. Therefore, we need to make the move to "standards-based testing".
The concept behind standards-based testing is simple. First of all, the classroom teacher presents the set of standards or objectives for which students will be responsible. In this manner, students know exactly what they are being tested on. They should no longer be able to say "I didn't know you were going to ask me that." In other words, there should be no surprises.
Secondly, the grading criteria are set in advance. When students are assessed, there are clear statements of evaluation criteria. Perhaps it is as simple as "This problem is worth two points - a score of two means you are correct, a score of 1 means you are not completely correct, but you have shown you know something about the problem, and a score of 0 means you did not demonstrate any knowledge about what was being asked." In other situations, for more complex tasks, a more detailed scoring rubric can be presented. The point is that students should know the criteria under which they are being evaluated. This also eliminates the issue of "scaling". When a student asks, "Are you scaling this test?" the answer is "No - you knew the objectives being assessed, and you knew the grading criteria. It was your responsibility to be ready for this assignment."
Thirdly, students are measured not against each other, but against these clearly stated performance objectives. Therefore, it is possible for ALL students to perform well. Indeed, it is the goal for all students to do extremely well. How many teachers have been told "Don't give out too many A's", or "How come no one received an A in your course?" When using standards based teaching and assessment, these questions can be answered confidently, by pointing to the standards and objectives, as well as the grading criteria, and showing that either the students have mastered these standards, or that the students did not master them.
It should not be the goal of teachers to give out 5% A's, 15% B's, 60% C's, 15% D's, and 5% F's. As teachers, it should be our goal to help ALL students achieve to a high degree. Standards-based instruction and assessment makes this possible.