UNC trying to slow grade inflation

by Lea Palmer, UNC sophomore, Carolina Journal, September 9, 2013.

Contextual grading may place spotlight on easy courses or lenient profs

In 2008, a committee of faculty members at UNC-Chapel Hill discovered that the average grade of a Carolina student was 3.213 — well over a B average. Led by Andrew Perrin, a professor of sociology, the school’s Educational Policy Committee began looking at how other universities were dealing with grade inflation and decided that “contextual grading” was the best way. Last fall, the university began to use it.

Contextual grading is the policy of reporting not just what the individual student earned in a course, but also what the class average was, thereby providing the “context” for the grade. Getting an A in a class where almost everyone gets an A is not so much of an accomplishment as getting an A when most of the other students earned Bs and Cs.

Contextual grading may not end grade inflation, but it may help reduce the effects of grade inflation. Using contextual grading, anyone can look into whether a student’s GPA may have been built up by taking lots of easy courses where As are common, or resulted from taking challenging courses where grades represent different levels of achievement.

Grade inflation has been a feature of American universities since the 1980s, says Stuart Rojstaczer, a former Duke University professor of geology, environmental science, and engineering. He is recognized as an expert on grade inflation and has developed a grading database of about 230 schools.

Some universities already have adopted policies to combat grade inflation. For example, Princeton University set a standard that fewer than 35 percent of the grades in any undergraduate class should be As. Wellesley College has a similar standard.

Beginning in 2009, Perrin and a team of other faculty members began examining other schools’ policies, but they rejected the “so-many-As-per-class” approach. Such systems assume that grading should be the same across all disciplines.

After several years, Perrin, his team, and the Educational Policy Committee decided on contextual grading. Each student’s transcript now contains not only the grade that particular student received but also the average grade given in that course section.

When this policy was announced, students were less than enthusiastic, and some faculty members questioned whether grade inflation should be a major concern. UNC-Chapel Hill vice-chancellor and provost Bruce Carney told The Daily Tar Heel, “Yes, we give high grades at Carolina, but I’ve heard faculty argue that we have better students than at other places.”

Recently, however, faculty, staff, and students at UNC-Chapel Hill have been more supportive, Perrin said in an interview with the Pope Center. “Having Carolina known for quality education and rigorous grading is good for students once they’re out on the job market and competing for graduate schools and so on.”

Perrin also notes that the university has contacted a number of graduate schools and large companies to ensure that this policy does not hurt the prospects of UNC-Chapel Hill graduates. The responses the university had received from various institutions have ranged from “not particularly interested” to very positive, and none of the respondents has criticized the new policy.

Many faculty members at UNC-Chapel Hill hope this policy will encourage students to take more challenging courses, Perrin said. He thinks some students avoid math and science courses because they are graded more stringently than other courses, including some in the humanities.

Faculty members also hope that adding more information to students’ transcripts also may prevent students from feeling entitled to a certain grade in a seemingly trivial course. Such entitlement is visible, for example, when a student complains to a professor about receiving a B in a course he or she thought would be an easy A.

Contextual grading is still in its early phases; the results of these first few steps will be analyzed after a five-year trial period.