Grade inflation describes an upward shift in the grade-point average (GPA) over an extended period of time and has emerged in all levels of education in the past three decades (Carney el al. 1978; Kolevzon 1981; Millman et al. 1983). The "average" grade of undergraduate courses is no longer a C, as suggested by most college catalogues. For instance, the average grade in all baccalaureate courses taught during the spring term at Pennsylvania State University rose from 2.67 in 1967 to 2.91 in 1974 (Nelson & Lynch, 1984). Furthermore, 83 percent of the grades given between 1992 and 1997 fell between A+ and B- at Princeton University (Elejalde-Ruiz, 1998). Even though the rise of grade can be associated with an equivalent rise in student ability is difficult to assess. According to the surveys conducted by the University of Washington's Office of Educational Assessment, 90 percent of faculty and a good portion of students think grade inflation is a problem and that discrepancies do exist in grade distribution among disciplines and types of courses. The erosion of grading standards has resulted in undesirable consequences for stakeholders, including the institution, employers, faculty, and students. These undesirable consequences stem from the ambiguity and non-comparability of grades within and among departments, schools, and universities, which make the grade interpretation difficult for stakeholders, i.e., cannot tell the differences between good work and very good work.
Grade inflation can generally be attributed to a number of factors. Becker (1975) and McKenzie (1975) propose theories which predict that teachers will inflate grades to achieve higher expected student evaluations of teaching (SETs). They give empirical support for this proposition. Nelson and Lynch (1984), and Greenwald and Gilmore (1997) also provide empirical results to support their conclusion: there is a positive correlation between a student's expected grade and evaluations given to instructors. Whether this correlation is due to easier grading or better teaching resulting in better grades and evaluations is controversial. While these relationships are difficult to determine, Worthington and Wong (1979) conduct a set of experimental studies of artificial grade manipulation, in which higher and lower?than?deserved grades are given relative to a control group. Their results indicate that students reward higher?than?deserved grades with better instructor evaluations. Other empirical studies by Stratton, Myers, and King (1994), Zangenehzadeh (1988), and Greenwald and Gilmore (1997) conclude that grading is influenced by SETs. It is generally agreed that, even without empirical studies to support the correlation between grading leniency and student evaluations, there is an inflationary bias in grading as long as instructors believe that students will give more favorable evaluations when their expected grades are higher.
Grade inflation may also be due to the desire of instructors to be popular among students. Or grade inflation has possibly been used as a way of price cutting when facing decline in demand for higher education. To maintain enrollments, educational institutions, rather than cutting their fees and tuition, have lowered the efforts associated with grades. So higher grades are used to offset the negative effects of the rapid increase in tuition. However, the higher grades are not adjusted back to their previous level when the demand for higher education rises.
Goldman (1985) gives a comprehensive discussion on the causes for grade inflation. In addition to the enrollment patterns and student evaluations of faculty performance factors, he also discusses a number of other factors. They include the Vietnam War, the culture of narcissism, the philosophy of relativism, institutional innovations and changes in administrative practices, the nature of students, the nature of faculty members, shifts in gender and racial proportions in higher education. …