Measuring Up: Using Test Scores Correctly
Here’s the scene: you go to the mailbox and find in the day’s delivery your child’s scores from the Iowa Tests for Basic Skills.You open them as you walk back to the house, and you see with great joy that she is doing splendidly in math. But—you stop walking—the verbal skills are only 38% and the science score even worse. A pang goes through your heart. You ask, "Am I doing a bad job?"
Perhaps. But the paper you hold in your hand is not the way to know that. The scores are not telling you anything about yourself or your curriculum. They are telling you only about your child’s performance on one particular day in some subjects.
There are two major kinds of standardized tests: aptitude tests and achievement tests.
Standardized aptitude tests predict how well a student may perform in some subject or skill in the future.
Standardized achievement tests try to provide a valid comparison of a student with a national sample of students of the same grade level.This is a big job.
Consider some of the challenges: to test a student’s knowledge accurately and fairly, and within a reasonable time limit; to determine a national norm that really represents the nation; and to provide a good tool to help identify a student’s relative strengths and weaknesses.
For the most part, the standardized tests do this job well. So far so good. But the trouble comes when people misuse the tests, when they expect the tests to tell them more than they were meant to tell. Often people use the tests to say that a teacher is doing a good job or a bad job. But that judgment would require an entirely different kind of assessment.
In his article "Why Standardized Tests Don’t Measure Educational Quality," W. James Popham (a report from the Association for Supervision and Curriculum Development) says, "Employing standardized achievement tests to ascertain educational quality is like measuring temperature with a tablespoon. . . . Standardized achievement tests have a different measurement mission than indication of how good or bad a school is. Standardized achievement tests should be used to make the comparative interpretations that they were intended to provide."
So why don’t standardized tests clue us in to how well a teacher is teaching?
For one thing, because tests have to be given all across the nation, they tend to include questions that are often quite general. The test makers have to accommodate the wide differences in requirements across the states. It would be impossible to make a test that perfectly suited every classroom in every state.
Another reason is that, in order to maintain this middle ground, tests do not sort out the natural abilities or the different kinds of intelligence in the test takers. Both of these considerations have a major effect on how well students do on the standardized tests, and neither is under the control of the teacher. Thus, to assess a teacher’s work by the test is inherently unfair.
Achievement tests can help you see where your child is doing well and where he is not. They can, over time and enough taking of them, show you improvement or decline in student performance. But they cannot—and must not be assumed to—indicate whether you are a good teacher. A naturally intelligent student with wide experience could conceivably do very well on such a test without any teacher at all.
So when those scores come in the mail, read them and use them as they were designed to be used—as a thermometer for the student’s achievement at that moment in time. If you want to know how you are measuring up, you will need to get a different instrument. Perhaps a coffee spoon (and a cup of coffee), over which you can talk with your family.