Beyond Grade Inflation: Grading Problems in Higher Education (review)

Beyond Grade Inflationcan be regarded as an extended review article which surveys recent issues on the topic of grade inflation. Among its important contributions are a discussion of demographic variables thought to be associated with recent trends in the distributions of college grades, evidence supporting the very existence of grade inflation, potential causes of grade inflation, and possible remedies for it.

In addition to providing a useful review, the primary innovation of this book might be Shouping Hu's disaggregation of grade inflation into a number of constituent pieces. These include grade increase, grade inflation, grade compression, and grade disparity. Grade increase refers to the (possibly) increasing trend in college grades over time, while grade inflation occurs if a grade increase is unwarranted by higher student performance.

Grade compression is a consequence of increasing grades, which presumably makes it difficult for faculty to accurately report differences in student performance at the higher end of the performance spectrum. When, for example, 30% of students get A's, how does one report the performance of a truly outstanding student?

Grade disparity refers to the fact that different professors often award drastically different grades for what is essentially the same level of student achievement. A primary conclusion of the book is that grading disparity "is a serious threat to the integrity of college grading. It not only affects students' choice of courses but also provides incentives for faculty to lower their grading standards" (pp. 42-43).

The book correctly identifies several causes of grading disparities and grade inflation. "Among all the suggested causes, student evaluation of teaching has become one of the primary reasons for grade inflation in colleges" (p. 55). A perception exists among many college faculty members that students who get high grades return favorable course evaluations. However, the book does not go far enough in recommending plausible remedies for correcting either grading disparities or the misinterpretation of student evaluations of teaching. Relatively simple solutions exist to both problems. Here is one of many not discussed in the book: Selectively exclude student evaluations of teaching based on observed grade distributions.

To see how this idea works, suppose that an institution agrees to an upper boundary of 30% A's, and suppose that a particular professor teaches a class of 100 students and awards 32 A's. Because this professor awarded two A's over the limit, the two best student evaluations of teaching would be excluded from summaries of that professor's teaching assessment.

On the other hand, there is also a perception among professors that students who earn low grades tend to submit negative student evaluations of teaching, thus discouraging the assignment of C's, D's, and F's and indirectly fostering grade inflation. To counter this notion, faculty should be permitted to exclude one negative student evaluation of teaching for every C, D, or F that they award. By itself, this simple selection process would largely eliminate incentives derived from student evaluations of teaching for faculty to inflate grades. It would also significantly alter classroom dynamics.

Much of the book is devoted to a discussion of whether grade inflation or grade increase has occurred. Central to this discussion is a detailed examination of student and faculty demographic variables that might explain grade increases observed at many colleges, particularly at the elite universities.

In my opinion, this discussion completely misses the mark. Whether student and faculty demographics have changed is irrelevant to the question of whether grade increases have occurred. College grades are subjective measures of student performance, and anyone who thinks that there is a historical anchor upon which a grade of A (or B or C) can be chained simply doesn't understand college grading practices or the purpose of college grades.

College grades represent a comparative measure of student performance, and the relevant comparison groups are student contemporaries. It is essentially impossible to compare the...