AS THE ADMISSIONS OFFICE PROUDLY reports, the student body at Duke has never been better. The university now competes with the Ivies for the nation's elite high-school seniors. These kids are so bright that they can correctly spell Krzyzewski and Kierkegaard--and summarize the lifetime records of both. But are they so uniformly excellent that most of them are doing grade-A work? The average grade in the average course at Duke is now approaching A-minus--and, if anything, rising. When A stands for average, some faculty members are now asking, do grades mean anything at all?
How bad is grade inflation? Think of it this way: grades seem to have risen almost as fast as tuition. National research by Arthur Levine, now president of Teachers College at Columbia University, shows that students averaging A-minus or better rose from 7 percent in 1969 to 26 percent in 1993. C performers became as rare as the old elite, falling from 25 percent to 9 percent. Inflation--and the search for a cure--seems most acute at some of America's finest colleges and universities. Two years ago Stanford brought back the F, retired in 1970, after A's and B's accounted for 93 percent of all grades. Dartmouth and Williams have also tinkered with their systems to make transcripts more meaningful to graduate schools and employers. And Harvard (where 80 percent of students now qualify for ""honors'') is studying reform. But no idea rivals Duke's for radical change. The school is considering a complex plan to eliminate the standard grading system and replace it with a formula that will make grades a matter of campuswide, interdisciplinary comparisons. ""It's very ambitious,'' says Ramon Saldivar, dean of undergraduates at Stanford. ""Everybody's going to be watching.''
Duke reformers think the GPA has become so badly overweight that the only solution is death--not a diet. ""Even B-minuses are starting to raise eyebrows,'' laments Valen Johnson, the statistics professor who created Duke's alternative to the GPA (hopefully dubbed the ""achievement index''). A vicious circle is at work with the GPA, Johnson argues. Teachers can ladle out high grades without having to discriminate much among their students. Students don't complain, and neither do their parents: for $1,000 a week, they figure, Junior ought to bring home some snazzy grades. All that's lost is judgment about the comparative value of student performance.
One problem with Johnson's index, the ""AI,'' is that even he admits he can't explain it simply. His defense: he can't easily explain the Dow Jones industrial average or the consumer price index, either. Professors would grade students in the usual way, but the results would then be fed through a computer to adjust for levels of difficulty. The index relies on an algorithm that measures one student against real-world classmates--not a subjective 4.0 ideal. It rewards those who do well in classes with a wide distribution of grades. Earn a B-plus in an economics class where the average grade was B-minus, and your AI rises. Earn an A-minus against everybody else's A in an art seminar and it falls. The best analogy comes from sport--the computerized rankings of pro golfers and tennis players, or even college basketball's Top 25. Coach Mike Krzyzewski's team ranked sixth in the nation last week despite five losses, notes Johnson, ""because we play some tough schools. The same should apply to classes.''
On March 13, a 40-member Duke faculty committee will decide whether to try the index experimentally by fall 1998. It looks like they'll vote yes, narrowly. But students, who will hold a referendum late next week, overwhelmingly oppose the AI. ""At least 70 percent are against it,'' says student-government president Takcus Nesbit. Among the complaints: the index would make competition more cutthroat and leave their futures to the mercy of an arcane formula no other college uses. Some faculty even make the case that the AI's dishonest--that it aims to quantify the unquantifiable for the crude convenience of professional schools and corporations. Quality isn't always relative, argues art-history professor Annabel Wharton. ""A great Manet is not better than a great Rembrandt.'' True enough. But is every kid at Duke really an old master?