As human beings we have the need to be able to evaluate our own intelligence and progress as we grow and participate in the educational process. We do this through various means. The most common way we do this is to test our youth at various ages and grade levels. We decide what topics to include on the tests, what kinds of questions to ask, what the correct answers are, and how to use the resulting test scores. These tests can be made in two different formats, Norm-referenced tests and criterion or standards-referenced tests. This paper will describe each, their limitations, their uses, how they affect the schools and students who take them, and whether or not they are still useful in educating today's students.
Norm-referenced tests compare a person's score against the scores of a group of people who have already taken the same test, called the "norming group." Scores are reported in percentiles, quartiles, stanines, and/or grade equivalents. Norm-referenced tests are designed to compare students" scores. Commercial norm-referenced tests do not compare all of the students who take the test in a given year. Test-makers select a sample from a given student group, i.e. all 6th graders. The students in the target group take the test. The test is then "normed" on this sample. It is supposed to represent the entire target group in the nation. Students" scores are then compared and reported in comparison to the norming group. .
The test makers try to create exams that provide results that, in the end, look like bell-shaped curves. This is to make comparison easier. In this scenario, most students will score near the middle, a few will score low, and a few will score high. In order to achieve the tests sometimes will emphasize meaningless differences among the students. These tests are designed to rank and sort students so many times "common knowledge" type questions are not used.