Michael Braun's Blog

I’ve been pondering this article in the New York Times since I read it yesterday morning. It echoes a frequent concern among older adults that today’s school grading standards are not tough enough. Subsequently, there is rampant grade and ego inflation. Students have no trouble getting A’s in high school and expect similar in college. When they don’t get A’s, they complain and make trouble for professors. When they do get A’s, it serves as proof of coddling to those concerned about weak standards.

What’s their solution to this problem? Institute a brutal curve, cutting students down to that classic bell shape. Only the top students get A’s. Mode, Median, and Mean all fall smack at C. And no matter the performance, some students will fail. It’s easy to critique the enforced curve on the basis of emotion. How cruel to punish those average students with C’s! To proponents of the curve, this serves as further reinforcement that it’s the right thing to do.

Examining the difficult task of test construction reveals that a brutal curve may not be the best solution to grade inflation. Consider the following scenario. A university department institutes a mandatory brutal curve solution. All scores must be adjusted to have their central point fall at C. Thus, the department is forced to mandate that all test scores be easily quantifiable. There must be no subjectivity in grading. If there was, students would likely be against the system. The department also standardizes curriculum and tests, to ensure that students are tested as fairly as possible.

What would the department find? After a year with this policy, they are likely to see that their tests do less to measure student learning and more to measure instructor skill. In a course with multiple meeting times and professors, Professor Alger inspires her students to learn. She teaches with vigor and passion. Her average test grade is an excellent 88%. Unfortunately, for her students, their 88% is not a B, it’s a C. Only students scoring above a 97% earn an A. This is completely different from Professor Franco. He’s close to retirement and is not keen on the university’s new standardized curriculum. Thus, he continues to lecture from his 25-year-old outlines. Students are bored and confused, but he offers no help. His average test grade is a dismal 45%. Any student scoring above a 70% gets an A.

With results like this, who would argue that the department should keep their brutal curve? This type of grading punishes good teachers and rewards bad ones. It accommodates poor test writing, uninformative lectures, and intense competition. It offers no benefit to the creation of a community of learning. It gives students no incentives to discuss material outside of class or ask questions during lecture. After all, all students should perceive each other as enemies in this zero-sum game.

The problem with grade inflation has nothing to do with the addition or subtraction of a brutal curve. It has to do with teachers who are unable to craft exams that test their students’ knowledge. Each examination should be designed to separate students who know the material from students who do not. This level of difficulty should be measured against the percentage breakdowns for each letter grade. Using a curve may give a general separation of those students who know the material and those who do not, but it’s like using a cleaver to fillet a fish.

I agree that universities who regularly hand out 50-60% A’s are doing something wrong. But making the examinations harder is a much better solution for testing knowledge than is a brutal curve. Making grading itself harder does nothing to increase a student’s knowledge. That’s the teacher’s job, not the grading scale.