by Dr. Mark H. Shapiro
"What are the characteristics of a good grading system?
• It should be rigorous, accurate, and permit meaningful distinctions among students in applying a uniform standard of performance.
• It should be fair to students and candid to those who are entitled to information about students.
• It should be supportive of learning and helpful to students in achieving their educational goals.".... ...Henry Rosovsky and Matthew Hartley, Evaluation and the Academy: Are We Doing the Right Thing? (American Academy of Arts and Sciences, 2002)
Commentary of the Day - November 18, 2002: The Dangerous Myths of Alfie Kohn.
We have commented frequently on the issue of grade inflation. Most recently, The Irascible Professor highlighted an extensive report on grade inflation that was issued by the American Academy of Arts and Sciences earlier this year. This report, which was authored by Henry Rosovsky, former Dean of Harvard's Faculty of Arts and Sciences, and Matthew Hartley, a lecturer in the Graduate School of Education at the University of Pennsylvania, was notable both for its detailed discussion of the available data on the phenomenon and for an extensive examination of both the causes and consequences of the practice. The IP's commentary addressed both of these issues in some detail.
Meanwhile, author Alfie Kohn who has published books such as No Contest - The Case Against Competition; Punished by Rewards - The Trouble with Gold Stars, Incentive Plans, A's, Praise and Other Bribes; The Schools our Children Deserve - Moving Beyond Traditional Classrooms and "Tougher Standards"; and The Case Against Standardized Testing - Raising the Scores, Ruining the Schools, among others, has weighed in on the issue in an essay published in the November 8, 2002 issue of The Chronicle Review entitled "The Dangerous Myth of Grade Inflation". From the titles of Kohn's books, it's clear that he is not a person who has much use either for standards or for competition. Indeed, the No Contest book is something of a polemic against all forms of competition in society, not just competition in school settings.
Thus, it comes as no surprise that Kohn is particularly troubled by recent suggestions that grade inflation is something to worry about. However, the thrust of his argument in "The Dangerous Myth of Grade Inflation" is more than a bit curious. First he argues that there really has not been all that much grade inflation in the past few decades, but then he argues that if there is grade inflation it's not anything to worry about because grades are intrinsically bad to begin with.
Kohn makes four arguments against the use of grades in college courses. First he argues that it should not be the professor's job to sort students for employers or graduate schools. Second he argues that students should not be set against one another in a race for the artificially scarce rewards of high grades. Third he argues that lower grades do not imply higher standards. And, finally he argues that grades do not motivate. Kohn, feels that these basic issues should be at the heart of any discussion of grade inflation.
The IP thinks that Alfie has set up a series of straw men, and that both his facts and logic are faulty. Let's examine each of Kohn's arguments in turn. Kohn's first contention is that some in academia are concerned about grade inflation (or compression as Kohn terms it) because it makes it "harder to spread students out on a continuum, ranking them against one another for the benefit of post college constituencies." Kohn asks if it the professor's job to "rate students .... for the convenience of corporations, or to offer feedback that will help students learn more skillfully and enthusiastically?"
This seems to the IP to be the epitome of a false dichotomy. Professors do both. The grading process, particularly the formative kinds of grades that are given during the term on quizzes, homework, and midterm exams are there for the purpose of providing feedback to the student about how he or she is progressing in the course. And professors frequently provide information to prospective employers about students. But, based on his 32+ years of teaching the IP can assure Kohn that few decisions about corporate hiring are based directly on grades. While most prospective employers expect students to present a reasonably decent transcript, it seldom is the case that one candidate will be chosen over another simply because his or her overall grade point average is a few tenths of a point higher. Prospective employers usually are much more interested in the results of interviews, letters of recommendations, or the results of telephone conversations with the candidate's teachers. The IP has supplied scores of letters for students and also has talked with numerous corporate personnel officers about students who have given his name as a potential recommender. In not a single case has the IP been asked a specific question about the candidate's GPA. Instead, the questions more often focus on the strengths and weaknesses of the candidate for the particular job.
The IP also has filled out hundreds of recommendation forms for students seeking admission to graduate programs. Naturally, graduate admission committees are interested in a student's relative strength among his or her peers. But again, the forms usually request qualitative judgments not only about the student's relative rank, but also about many other characteristics that the committee thinks might be important. In addition, the committees almost always ask for a narrative letter describing the candidate's strengths and weaknesses. Clearly, grades are an important factor in graduate and professional school admissions, but they are hardly the only factor. Medical schools, for example, often are as much interested in character issues as they are in grade issues.
Kohn's second argument against the use of grades apparently stems from his strongly held belief that competition is inherently bad, and that grading "on the curve" ensures that only a few students will be able to get A's no matter how well that the class as a whole does. According to Kohn: "A bell curve may sometimes ..... describe the range of knowledge in a roomful of students at the beginning of a course. When it's over, though, any responsible educator hopes that the results would skew drastically to the right, meaning that most students learned what they hadn't known before."
Again Kohn gets it wrong. In fact, he gets it wrong on two counts. First, he assumes that university instructors base their grades entirely on test results that follow a more or less normal distribution. In the IP's experience this is seldom the case except perhaps in very large introductory course. More often than not grades are based on a number of factors both quantitative and qualitative. These include quizzes and tests to be sure, but they also include such things as class participation, homework, term papers, etc. Second, he assumes that all students will achieve the same gains in knowledge and understanding as the course proceeds. Most of us who teach wish that this were so. But, we know that learning involves a partnership between instructor and student. If both instructor and student are doing their jobs, learning occurs. But even if both approach the task with the best of intentions, the outcome is not always the same for every student. Some students are more capable than others, some have more time to devote to the course than others, some come to the course better prepared than others. At the end of the course it certainly is true that the mean level of knowledge and understanding has improved, often substantially. However, there still usually is a distribution in performance. Some students will have achieved more, some less. The final distribution in a course seldom is "normal". Instructors generally use a good bit of judgment in interpreting that distribution in an attempt to ensure that students have been fairly graded.
Second, Kohn gets it wrong when he assumes that students are "set against one another in a race for artificially scarce rewards". He creates the impression that the desire to get good grades turns students into cutthroat competitors, the stereotypical "grade grubbers" that he refers to in his article. However, in reality, this seldom is the case. To be sure students like to get good grades, but in his more than 32 years of teaching the IP has run into only a handful of students who seemed to be obsessed with grades in the sense that Kohn describes. College is not an athletic competition where only a few come away with the medals. Study groups and collaborative effort in preparing for exams is the rule rather than the exception among college students. The competition for grades is much more in the nature of the friendly rivalry described by Gordon, than the high stakes race described by Kohn.
Kohn's third argument about the relationship between grades and academic standards is perhaps the most confusing of his four points. He seems to equate grades with standards in the sense that professors who feel that they have high standards tend to give lower grades, while professors with low standards tend to give high grades. Kohn takes some decidedly cheap shots at faculty members who lament grade inflation, describing them as cranky old professors who really regard teaching as "an occupational hazard" to be avoided rather than engaged in. He seems to imply that any instructor who does not give every student in his or her class a top grade is an uncaring lout who is not interested in the welfare of students.
The IP has encountered a few instructors who fit Kohn's description, but they are the exception rather than the rule. Most instructors really do care about their students, and they do not take any great joy in having to give a student a low grade in a course. But at the same time, it ought to be obvious that they do have a right to expect some minimum level of performance before awarding a passing grade to a student. Come on Alfie, wouldn't you be at least a little concerned if you were driving across a bridge designed by an engineer who got the "lowest A" in a class where only A's were awarded.
Kohn's fourth argument that we should eliminate grades because they don't motivate also is specious. Grades are not meant to motivate. They are meant to inform. Here Kohn confuses the motivational effects of tests with the motivational effects of grades. Certainly, some students are motivated to study only because there is the prospect of an exam, term paper, lab report or other exercise on the horizon. Some students may be motivated by the desire to do well in a particular course or to get high grades in general, but most are motivated by other factors. Kohn worries that grades create an extrinsic reward structure that is "likely to undermine the love of learning we are presumably seeking to promote". Again, based on his 32+ years in the classroom, the IP would disagree. The factors that motivate students are many and complex. Some are extrinsic and some are intrinsic. Probably the majority of students who attend large public comprehensive universities like the one where the IP teaches don't enter with the intrinsic love of learning that Kohn speaks about (this may not be true for students entering more elite institutions). They are more likely to be motivated by learning as a means to an end. Nevertheless, some do bring to their university experience a love of learning, and many additional students develop a love of learning along the way. This has little to do with grades.
One of the most rewarding things about teaching physics has been the opportunity to interact with physics majors. The majority of these students have a genuine love of the discipline. That love of the discipline is as strong among the "average" students (those who generally receive C grades) as it is among the outstanding students.
The greatest danger from the myths that Alfie Kohn propagates is that they deflect attention from more important questions such as how to ensure that our grading procedures are both fair and effective -- fair both in the sense that they provide a rich range of opportunities for a student to demonstrate learning and fair in the sense that the playing field has been leveled as far as possible -- effective in the sense that the process provides timely information both to the instructor and to the student.
© 2002 Dr. Mark H. Shapiro - All rights reserved.