By Gary Deel, Ph.D., JD
Faculty Director, School of Business, American Public University
Throughout my undergraduate years, the courses I took were never terribly challenging for me. This is not to brag. I didn’t major in nuclear physics; my bachelor’s degree was in hospitality management.
Start a degree program at American Public University.
I cared about the quality of my work, but I never felt particularly compelled to finish at the very top of my class. I like to think that I could have done so if I had felt inclined.
But earning high Bs and low As seemed almost effortless, and I was content with that. Then I went to law school and everything changed.
My First Exam in Law School Was an Eye-Opener
I’ll never forget the first exam in my first class. The test scores were posted anonymously using our numerical student IDs.
I went up to the wall with the rest of my 90 or so classmates in the night program and looked for my number. I felt anxious; it was like being back in high school and searching through the newly posted roster to see if I had made the school soccer team.
I finally found my number and my score. 14. “…out of what?” I thought.
I asked a classmate. “Out of 100,” he said.
What a humbling experience that was. Over the next four years, I came to learn how law schools engineer assessments to push the limits of student potential and produce brilliant critical thinkers. Here’s the general idea.
Rarely Does a Student Actually Earn a Passing Grade from the Raw Scoring
Every assessment in every class is designed to be extremely challenging, so much so that rarely does anyone actually earn a passing grade from the raw scoring. Even the brightest and most well-prepared students are taken to their breaking points.
But the raw scores are not the final scores. Thereafter, the professors shift the bell curve of grade distributions upward. The highest grade — even though it might be a failing grade to begin with — becomes an A, and every other grade falls in sequence below that A, with the same spread and standard deviations as the original scoring.
This is commonly referred to in academia as grading on a curve, and it is controversial to say the least. But having experienced it as a student, having used grading on a curve for nearly a decade as a professor, and having heard the worst criticisms of this assessment method, I nevertheless advocate grading on a curve throughout higher education.
Here’s why. I’ve designed and redesigned many courses for many schools during my academic career. And whenever we professors design an assessment (an exam, a written assignment or other gradable task), we hope that it will measure what we want to measure. We also hope that it will test students at the limits of their knowledge and abilities. Finally, we hope that it will accurately reflect their competencies.
But it doesn’t always work out that way. For example, suppose we have a newly designed or redesigned test, and we roll it out to a new class of students. The average comes back an A-. One-third of the students earned perfect scores. Most of the rest earned grades of B or higher, with only one or two outliers in the C and D range.
We Obviously Haven’t Fully Challenged the Capabilities of a Huge Portion of the Class
Students who earned perfect scores are essentially communicating, through their performance, that the test was too easy for them. And in such cases, we do our students a disservice by not challenging them to their fullest ability and not pushing them to their highest potential.
Imagine you paid for a college course in literature. To your surprise, the curriculum turned out to be fourth grade-level reading comprehension. How might that make you feel? You’d likely feel frustrated and ripped off. You obviously mastered fourth-grade reading comprehension probably in fourth grade, or you wouldn’t be reading this article now.
What you ostensibly paid for was for the school and your professor to teach you something that you didn’t already know. They were supposed to challenge you, so that you would get the most out of the course that you possibly could for your personal and professional development.
Of course, this is an exaggerated example for illustrative purposes. But the same principle is true at any level. Anything short of optimal challenge and development is lost opportunity and lost value for the student.
But what if the opposite scenario occurs? Let’s suppose you roll out that new test, and the average score comes back a C-. One-third of the class failed. The highest grade — earned by just two students — was a B.
The problem with this scenario is that we have developed an assessment from which a perfect score should be possible, and a respectable score should be typical. Yet none of the students we subjected to the assessment was able to earn a perfect score. And so we’ve created a test that has proven to be so difficult that it is unfair.
Students are not professors. That is precisely why they are students. If we, as professors, set them up to fail with impossibly hard exams and give no consideration as to the level of challenge, then we have effectively rigged the game.
These kinds of assessments could be appropriately called “unicorns,” as they are nearly impossible to conquer no matter how much effort is committed. It should be quite clear that without intervention these assessments are objectively unfair.
So what can we do about it? We could certainly revise the test for future classes, and perhaps we should. Perhaps there are poorly worded questions, confusing instructions or curriculum alignment issues.
We also might want to finesse the exam to try to lower the difficulty slightly. However, we then run the risk of encountering the first scenario: an abundance of perfect scores indicating insufficient challenge.
Of course, regardless of how we modify it in the future, this plan does nothing to help the students who just failed the poorly designed exam. What about them?
Grading on a Curve Allows an Exam to Be Sufficiently Difficult to Challenge All Students
In my opinion, there is only one solution: grading on a curve. This approach allows an exam to be sufficiently difficult to challenge all students in the class to their fullest potential, even the most talented. Yet it also allows us to maintain justice by establishing a threshold that is reasonable and reflective of what all stakeholders in the higher education process might expect. In this sense, grade curving allows for difficulty in exams but eliminates the “unfairness” element.
I grade on a curve in my classes. If there are one or more perfect scores to begin with, then no curve would be applied.
But if no student achieves a perfect score, then the curve is shifted upward such that the highest grade in the class generally becomes a perfect score or its equivalent. For example, if the highest grade was an 85%, 15% or so would be added to all scores under the curve process. I design my assessments to be very difficult. Time parameters are tightly controlled, and the questions are such that there are no “easy” answers.
Low grades are not unusual. However, these assessments ensure that all students are challenged to their limits. When the grading curve is applied, the final grades will still reflect the relative performance of the students in the class, but at a point on the scale that accounts for the difficulty of the exam.
In a subsequent article, I will review some of the most common arguments against grading on a curve, and why I believe that each of these arguments ultimately fails.
About the Author
Dr. Gary Deel is a Faculty Director with the School of Business at American Public University. He holds a JD in Law and a Ph.D. in Hospitality/Business Management. He teaches human resources and employment law classes for American Public University, the University of Central Florida, Colorado State University and others.