There are two kinds of standardized tests given to schoolchildren to measure their yearly progress. They are totally different in purpose and interpretation.
* NORM-REFERENCED tests are designed to compare a particular student with all other students in the same grade. When these tests were written, test designers gave a pilot test to thousands of, say, tenth graders. For each item, they noted how many tenth graders answered correctly. That’s how they figured out what the “average” tenth grader should know. Norm-referenced tests compare your child’s score to all the other children in the U.S. who are in the same grade. Plotting the scores makes a nice “bell curve” with the fiftieth percentile being smack-dab average, and the highest part of the curve. Half of the students score above fiftieth percentile, and half below. Few children score extremely high or extremely low. That’s what a percentile score tells you: how did my kid do compared to the rest of the kids in his grade? If your child is in the 75th percentile, he did better than 75 out of a hundred kids in his grade who took the test. This score compares well to percentages, but don’t be fooled by that. Norm-referenced tests do not have a pass-fail level. They only measure any given student against others in the same grade. So what you see is how well your child “measures up” against the average child or your sister’s child.
* CRITERION-REFERENCED tests are a whole different animal. They are designed to assess how much of a given block of information your child has mastered. The test designers decide what knowledge and skills students in each grade should master as a minimum standard. Then they test to see how much of it each child has learned. These tests compare your child’s knowledge to a body of knowledge, not to other students. In my state, all public-school students must take a criterion-referenced test at each grade level. Certain grades (3, 5, 8, and 10) are crucial: a student must pass the test to progress to the next grade. If a certain score is not attained, the child is retained. So there is a definite pass-fail level in criterion-referenced tests.
Each kind of test has its own purpose, and schools are often woefully negligent in educating parents about them. This only adds to the stress that many parents and children experience at test time.