It is rare that chairmen of English departments become front-page fodder, but for a couple of weeks in February my phone rang with requests for interviews over an issue that seemed to me rather mundane. The dean of the college at Princeton University had released a report, on behalf of a university committee, that made an unsurprising observation: Grades were high, and were distributed among only a few options (A, A-, B+). The report concluded with the modest proposal that faculty members throughout the university gather in departmental meetings to talk about the implications of this situation.
Immediately, the undergraduate newspaper declared war on the dean, firing off salvos misquoting her as “demand[ing] immediate efforts to roll back the trend” of grade inflation. The student government voted to establish an undergraduate version of the university committee, and to compel faculty members to meet jointly with students on all university-wide discussions of grading policy. And The New York Times, as well as local newspapers, sent reporters scurrying to ask: Do professors feel pressured to give high grades? Are students smarter than they were 30 years ago? What is your department planning to do? As chairman of the largest humanities department, I was in special demand.
Frankly, much of the turmoil seems (dare I say it?) a tad inflated. After all, grading is among the more powerful gestures that teachers make and, at the same time, among the least examined collectively. I know of no university where new faculty members are uniformly counseled on the protocols of grading, or where colleagues regularly strive for consensus about what a particular grade means, or where deans customarily take the trouble to suggest that departments review their grading policies. Once a year, graduate students may be drummed through sessions that last a day or two, in which anonymous student essays are shared for mutual comment, followed by a generally desultory conversation about what levels of work merit particular grades.
But in a quarter-century of teaching since my own such graduate-school session, I’ve rarely been asked by colleagues what I meant by an A paper, or a B exam, or a C final grade. (Distressed students are another matter, of course, which is why I keep a box of tissues on my desk.) No one has ever suggested that my colleagues and I might want to consider discussing our expectations as a group, to transform grading into something more than a happenstance collection of idiosyncratic evaluations. The fact is that our Ph.D.'s and competence in certain scholarly materials authorize us to assign whatever grades we think students deserve, no questions asked. So the dean’s suggestion that departments consider the historical patterns of their grading is long overdue.
Defenders of the current state of grade inflation protest that students actually are stronger now; that lowering grades will hurt their chances for jobs or graduate school; that private universities owe high grades to high-paying consumers; or that elite institutions select only A students. But grades should be a shorthand for genuine discriminations among individual student efforts; all other considerations need to be secondary. It is always possible to indicate that a student, no matter how excellent in other ways or other courses, is better or worse than a classmate in meeting the demands of a particular course. No teacher is helped in making significant discriminations by limiting the range of possible grades to only two or three high categories. In my department, for instance, more than 75 per cent of all grades are B+ or higher. And I suspect that we are not alone, either among departments at Princeton or among colleges and universities nationwide.
I suggest that high, compressed average grades tell only part of the story, and the most benign part at that. Far worse is the problem of grade variation within and among individual courses, which is camouflaged by the bold averages conveyed by news accounts. My department has compiled data bases for each teacher, each course, and each section, and the results offer startling proof that faculty members grade in idiosyncratic ways. Throughout their careers at Princeton, some faculty members have awarded an A or A-to 35 per cent of their students, while others have awarded such grades to 87 per cent.
Variation among grading styles is hardly news, of course. More troubling is the fact that good students in different sections of large courses have much greater or lesser chances of earning top grades simply because they drew this or that instructor. In required English courses for sophomores, students select the section they will enroll in based solely on their time schedules, yet the proportion of A’s awarded in different sections ranges from 22 to 67 per cent. In required junior seminars, the range is between 38 and 70 per cent. The fact that the sheer luck of the draw can lead to such different evaluations for undoubtedly similar performances constitutes the true problem of our current attitude toward grading, in which we rarely speak to each other about what we expect of our students.
It is a matter of record that A’s everywhere are more abundant than a decade ago, that they mean far less as indicators of student performance, and that students themselves are divided about this development -- caught between wanting better grades and wanting to stand out from their peers. Of course, it’s only natural to desire good grades; all of us want our efforts rewarded, our class performances praised. And resentment of strict standards is understandable. I remember, three decades ago, waiting to speak to my newly assigned undergraduate adviser, whose middle initial (C, for Clarke), emblazoned on the office door, had been embellished years before by a sharp minus carefully etched in the wood by some grade-conscious undergraduate. (That may be why I’ve never shortened my own middle name.)
Among the many reasons adduced for the rise in grades since the 1960s is the introduction of course evaluations, which allow students to “grade” their teachers. And different groups of teachers may fear a backlash if they try to impose stricter grading standards: graduate instructors eager to make a good impression, which they can convert into success in the job market; junior faculty members committed to establishing reputations for popularity that will add to their prospects for tenure; and senior professors simply wanting to enhance course enrollments.
Yet, despite wide belief to the contrary, there is no fixed correlation between grading and course evaluations, at least in my department. I can say this because, as chairman, I see all grades and all evaluations. Some high graders receive low evaluations, some low graders receive high evaluations, and some professors receive evaluations similar to the grades they give. One interesting exception to this pattern is that high-grading graduate-student instructors, arbitrarily assigned to sections, are particularly likely to receive low evaluations. The only explanation for this that makes sense to me is that students are quick to recognize lack of pedagogical rigor, whether in grading or class preparation -- and while thankful for the former, nonetheless resent the latter.
Although grade inflation is troubling, I am opposed to any unilateral correction of the problem. We have arrived at the current situation over the course of many semesters, and any quick fix to the status quo will lead to such heated student protests that the goal of grading will be forgotten. By the same token, however, I want my department to resist any further rise in the grades we give.
How is that to be done? My faith in rational exchange on the subject, after two years of patient consultation with colleagues, has been sorely tested. At the beginning of each semester, I’ve e-mailed every teacher in the department, from the newest graduate instructors to the most experienced of senior colleagues, asking that everyone keep in mind our department’s recent history of awarding an overgenerous proportion of A’s. I remind them that we have agreed to correct this exaggerated high. I have also presented each colleague with a printout of his or her own historical grading average, as well as departmental grade averages for each course and level of course. When final grades are submitted, late in the semester, I’ve met with those whose grades fall well above or well below the department’s average, to ask them certain questions: Are the students in this class really so fine or so bad? Are you aware of the average grades given by other instructors in this same course? On what basis are you making your distinctions among students?
Occasionally, teachers with a reputation for low or moderate grading will have a class full of superlative students, and the average grade for the class will rise. But more often, grading follows a predictable pattern, with high grades continuing to be awarded by those historically inclined to bestow them.
Still, the message has been heard; although no radical change has taken place, there already seems to be less wild variation in the average grades awarded by various faculty members. And with fewer very high course averages, the overall average is slowly sinking.
Discussion alone is insufficient to alter the problems of grade inflation or wide variation in grades. Perhaps it’s because old habits die hard, and, at a time when faculty members in departments disagree about so much else, it is unlikely that they will readily alter a practice so central to their teaching. But it is no less true that, after decades of polite silence, a necessary first step must be to convene faculty members to discuss grading. We all need to think seriously about how to make our overall grading practices more equitable and more discriminating. Any such discussion can be only an improvement, after years in which we’ve each simply done what we thought was passing fair.
Lee Clark Mitchell is a professor and chair of the department of English at Princeton University.