The independent news organization of Duke University

Grading fail

Imagine if Duke’s basketball team was allowed to pick the teams it played, and wins and losses against teams of any caliber were weighed equally.

We’d fill our schedule with games against pathetic, small programs like Elon and Coastal Carolina (and Maryland), and go undefeated each year. This would be fun to watch, but also a pretty asinine way to run a basketball league.

That is why I am dumbfounded that a relatively smart bunch (that is, just about every academic dean in the country) have designed analogous systems for grading in universities. Students choose courses that vary hugely in average grade given—some professors give 50 percent of the class A’s, but others hardly give any. And in the end, each class is represented equally on a transcript, leaving job recruiters and anyone else to guess how a student really performed relative to his or her peers. No one knows the scale used when the classes were graded.

Beyond the unfairness inherent in this system, this situation presents serious problems for Duke, and even for our country as a whole.

For one thing, humanities classes are graded more leniently. According to a study released by retired Duke professor Stuart Rojstaczer, science classes at Duke and nationally tend to give grades that are, on the average, a whopping 0.4 grade points lower than those of their humanities counterparts. Another report on grading inequities at Duke compiled by Valen Johnson, a former professor of statistics, concluded that “differences in grading practices have a substantial effect on student enrollments, and cause fewer students to enroll in those fields that grade more stringently.”

In other words, at a time when our nation is in desperate need of science and math gurus, the predominant grading system incentivizes students to avoid these fields like the plague. Brilliant. At a more philosophical level, it’s simply wrong for any educational institution not to encourage students to challenge themselves. Career-oriented students look for “safe” classes where they won’t end up with an unexpected B-, often instead of the classes that naturally suit their interests. This is the natural reaction to a grading system that urges cautious souls to take the safer route and pick professors known to grade leniently.

This is hardly the first time this issue has been raised. In 1997, Professor Johnson proposed a new grading system called the “Achievement Index” that would have ranked students according to their performance in each class against the median class grade. This system was complex, and amidst serious push back from opponents concerned about the level of uncertainty added by a new ranking system, the proposal failed. Student leaders and administrators promised to look at the problem of grade variability across majors and devise a better solution.

Unsurprisingly, nothing was ever done. Today, in the face of overwhelming evidence, some administrators have taken a step backward and are denying that a problem exists at all. When I emailed Lee Baker, associate vice provost for undergraduate education, he wrote me that “some of our most popular majors include biomedical engineering, econ, neuroscience and biology.... So we do have evidence that many students do not select majors on ‘easiness.’” This is a bit like saying that since many people choose to become teachers, low salaries for teachers are OK.

In the meantime, Duke has gone from being a leader on this issue to an observer. Cornell and Dartmouth have both introduced median class grades on official student transcripts. UNC plans to do the same next year. Princeton now stipulates the percentage of A’s each academic department can dole out.

To be certain, following suit with any of these systems would have its drawbacks: Opponents of median grades claim they are poor indicators in small classes or seminars and that they disincentivize students from taking classes with more experienced students, which often have high median grades. Faculty complain that Princeton’s system constrains their freedom to evaluate students fairly and account for natural variations in performance. Others say that such systems add an unwanted level of competition to the classroom.

These debates are important to have—in fact, in response to an email I sent him, Steve Nowicki, dean and vice provost for undergraduate education, wrote “this would be an useful subject for a student-faculty group to discuss.”

Absolutely. But these kinds of groups are notorious for saying a lot and doing nothing, so the group tasked with exploring this issue needs to be told that the status quo—marked by most of the very same drawbacks associated with a new system and none of the benefits—is not an option.

During the 1997 debate, many opponents of new grading said that Duke should not be the first to try a new system. Why not? If our school wants to consider itself an innovator in something other than raunchy PowerPoint presentations, we need to lead sometimes.

Last year, President Brodhead wrote the student body that “Duke’s best tradition is that it’s not stuck in traditions.”

So let’s do something.

Jeremy Ruch is a Trinity junior and is currently studying abroad in Brazil, South Africa and Vietnam. His column runs every other Monday.


Share and discuss “Grading fail” on social media.