by Thomas R. Guskey, from Educational Leadership, Vol. 69:3
1. Avoid Using Grades to Differentiate Students
This is one of our oldest traditions in grading. It comes from the belief that grades should differentiate students on the basis of demonstrated talent. Students who show superior talent receive high grades; those who display lesser talent receive lower grades….
We must answer one basic, philosophical question, however: Is my purpose to select talent or develop it?…. If your purpose as an educator is to select talent, then you must work to maximize the differences among students. In other words, on any measure of learning, you must try to achieve the greatest possible variation in students’ scores….
If, on the other hand, your purpose is to develop talent, then you go about your work differently. First, you clarify what you want students to learn and be able to do. Then you do everything possible to ensure that all students learn those things well. If you succeed, there should be little or no variation in measures of student learning…..
2. Avoid Grade Distributions Resembling the Bell-Shaped Curve
The normal bell-shaped curve describes the distribution of randomly occurring events when nothing intervenes. If we conducted an experiment on crop yield in agriculture, for example, we would expect the results to resemble a normal curve. A few fertile fields would produce a high yield; a few infertile fields would produce a low yield; and most would produce an average yield, clustering around the center of the distribution.
But if we intervene in that process—say we add a fertilizer—we would hope to attain a very different distribution of results. Specifically, we would hope to have all fields, or nearly all, produce a high yield. The ideal result would be for all fields to move to the high end of the distribution. In fact, if the distribution of crop yield after our intervention still resembled a normal bell-shaped curve, that would show that our intervention had failed because it made no difference.
Teaching is a similar intervention. It’s a purposeful and intentional act. We engage in teaching to attain a specific result—that is, to have all students, or nearly all, learn well the things we set out to teach. And just like adding a fertilizer, if the distribution of student learning after teaching resembles a normal bell-shaped curve, that, too, shows the degree to which our intervention failed. It made no difference.
Further, research has shown that the seemingly direct relationship between aptitude or intelligence and school achievement depends on instructional conditions, not a normal distribution curve. When the instructional quality is high and well matched to students’ learning needs, the magnitude of the relationship between aptitude/intelligence and school achievement diminishes drastically and approaches zero.
3. Avoid Basing Grades on Students’ Standing Among Classmates
Most parents grew up in classrooms where their performance was judged against that of their peers. A grade of C didn’t mean you had reached Step 3 in a five-step process to mastery or proficiency. It meant “average” or “in the middle of the class.” Similarly, a high grade did not necessarily represent excellent learning. It simply meant that you did better than most of your classmates….
But there’s a problem with this approach: Grades based on students’ standing among classmates tell us nothing about how well students have learned. In such a system, all students might have performed miserably, but some simply performed less miserably than others.
In addition, basing grades on students’ standing among classmates makes learning highly competitive. Students must compete with one another for the few scarce rewards (high grades) to be awarded by teachers. Doing well does not mean learning excellently; it means outdoing your classmates. Such competition damages relationships in school. Students are discouraged from cooperating or helping one another because doing so might hurt the helper’s chance at success. Similarly, teachers may refrain from helping individual students because some students might construe this as showing favoritism and biasing the competition.
Grades must always be based on clearly specified learning criteria. Those criteria should be rigorous, challenging, and transparent.
[Michael’s Note: This means we should not curve student grades on assessments.]
4. Avoid Using Poor Grades to Motivate Students
No research supports the idea that low grades prompt students to try harder. More often, low grades prompt students to withdraw from learning. To protect their self-images, many students regard the low grade as irrelevant or meaningless. Others may blame themselves for the low grade but feel helpless to improve.
Recognizing the effects on students of low grades, some teachers have initiated policies that eliminate the use of failing grades altogether. Instead of assigning a low or failing grade, teachers assign an I, or “Incomplete.” Students who receive an I may be required to attend special study sessions to bring their performance up to an acceptable level…. [In other words, if it’s worth assigning, it’s worth insisting that a certain level of mastery be attained.]
Guskey is a professor of education at the University of Kentucky and one of the foremost experts on assessment. For the research supporting this article, see: http://www.ascd.org/publications/educational-leadership/nov11/vol69/num03/Five-Obstacles-to-Grading-Reform.aspx
See also in ClassWise: