|
« AAUP Homepage
|
Grade Inflation: A Crisis in College Education
Reviewed by Raymond P. Perry
Valen E. Johnson. New York: Springer Publishers, 2003
Valen Johnson's provocative book is sure to arouse heated debate on two topics dear to the hearts of professors: grade inflation and student ratings. Johnson argues that grade inflation undermines the evaluation of college teaching, distorts course enrollments, and biases resource allocations, all to the disadvantage of the natural sciences. Johnson supports these arguments with data collected from a field study at Duke University, which required students to evaluate their professors and courses using a Web-based system. Initially planned to run for three years, the study was aborted after one year because of technical problems and opposition from faculty and administrators. Johnson's support of his claims is strikingly uneven, and he fails to present evidence to underpin such broad assertions as "the general level of scientific competence in America has been diminished simply because universities have not adopted more consistent grading policies."
The book's eight chapters are organized around three main assertions: grade inflation is a result of professors "buying" student ratings with course grades, grade inflation distorts course enrollments to the detriment of the natural sciences, and these enrollment differentials lead to resource allocations that jeopardize the natural sciences and, therefore, higher education generally. After a brief overview, Johnson describes his field study, which purportedly addresses the methodological and statistical flaws of previous research and provides a definitive analysis of grade inflation issues, the most vexing of which is the positive, though weak, correlation between grades and student ratings of instruction. Johnson concludes that higher student ratings are not simply correlated with higher course grades, but rather cause them.
In an overly selective literature review, Johnson attempts to demonstrate the threat that student ratings pose for higher education, focusing on the reliability and validity of existing ratings questionnaires used, types of experimental designs used to develop such questionnaires, and theories about the correlation between grades and student ratings. He assesses these theories using the data from his field study and quickly concludes that student ratings are invalid because professors grade leniently to obtain better student ratings. He rejects the theory that better professors produce higher grades, and that because of this, students award higher ratings—a theory that supports the validity of student ratings and that is endorsed by experts in the field.
Johnson then introduces what he deems to be two other detrimental aspects of grade inflation: course enrollment patterns and grading disparities between academic disciplines. He claims that students are more likely to take courses from professors who assign higher grades and finds that the natural sciences, mathematics, and economics are the most stringently graded courses and the humanities the most liberal, with the social sciences falling in between. Johnson applies an impressive array of statistical procedures to eliminate alternate interpretations, many of which require substantial conceptual and technical background to comprehend. In closing, he provides policy recommendations designed to counter the threats posed by grade inflation.
Johnson's effort to document the deleterious effects of grade inflation has several notable features. First, his writing style is lucid and the book reasonably well organized—both salient qualities of effective teachers. The systematic chapter development; the copious use of supplemental tables, figures, and appendices; and the detailed explanations of sophisticated statistical procedures make the book easier to comprehend than it might otherwise be. Second, Johnson brings complicated, disparate literatures to bear on grade inflation; his articulation of various theories provides a moderately useful guide in understanding the correlation between grades and student ratings. Third, his use of advanced statistical procedures helps rule out competing explanations and introduces innovative techniques not common in past studies. Finally, Johnson tests his hypotheses using an actual field study, flawed though it may be.
But serious conceptual and methodological problems plague the book. Johnson's analyses are based on a case study of a single institution that has unique attributes. Would the evaluation of teaching be so strenuously opposed in a liberal arts college as in a research-intensive university? Moreover, the newly developed survey was administered online, a relatively novel approach, but one that research does not endorse as superior to paper administration. Perhaps the fact that the survey was online explains why 71 percent of students did not participate—a substantial percentage when compared with established practices of evaluating professors and courses, though not compared with paper surveys in general.
Rather than adopt one of the widely used standardized questionnaires that exist, Johnson and his colleagues, who would not be considered experts in the field, constructed an entirely new form. The study may demonstrate that the Duke questionnaire itself was invalid, perhaps because of poor design and implementation, but it does not demonstrate that properly designed student ratings questionnaires are invalid. Moreover, Johnson's literature review downplays the empirical evidence supporting student ratings. For example, Johnson exaggerates the amount of evidence demonstrating the existence of the so-called "Dr. Fox effect," which claims that professors can obtain high ratings without teaching much content, if they are enthusiastic and entertaining. In fact, the bulk of high-quality studies do not support this conclusion.
Johnson believes that grade inflation will be the downfall of higher education and that student ratings are higher education's Achilles heel. Though grade inflation may be an insidious problem, Johnson does not make a definitive case that student ratings are the culprit. How does their importance compare, for example, to the more general degradation of faculty autonomy that must surely contribute to grade inflation? Perhaps Johnson has made a useful start in stirring up a debate about grade inflation and the validity of student ratings. But instead of concerned, but misguided, individuals reinventing the wheel every so often, perhaps it's time that a blue-ribbon committee, established by a consortium of our professional organizations, provide some definitive guidance on student ratings.
Raymond Perry is professor of psychology at the University of Manitoba.
|