Grades on the Rise

Grades awarded to U.S. undergraduates have risen substantially in the last few decades, and grade inflation has become particularly pronounced at selective and private colleges, a new analysis of data on grading practices has found.

In “Grading in American Colleges and Universities,” published Thursday in Teachers College Record, Stuart Rojstaczer, a former Duke University professor of geology, and Christopher Healy, an associate professor of computer science at Furman University, illustrate that grade point averages have risen nationally throughout most of the last five decades. The study also indicates that the mean G.P.A. at an institution is “highly dependent” upon the quality of its students and whether it is public or private..

“There’s no doubt we are grading easier,” said Rojstaczer, the founder of GradeInflation.com, where he’s built a database of grades at a range of four-year institutions since 2003. The findings are based on historical data dating back at least 15 years at more than 80 colleges and universities, and contemporary data from more than 160 institutions with enrollments totaling more than 2,000,000.

Since the 1960s, the national mean G.P.A. at the institutions from which he’s collected grades has risen by about 0.1 each decade – other than in the 1970s, when G.P.A.s stagnated or fell slightly. In the 1950s, according to Rojstaczer’s data, the mean G.P.A. at U.S. colleges and universities was 2.52. By 2006-07, it was 3.11.

Though there’s “not a simple answer as to why we grade the way we do,” Rojstaczer speculated on several reasons why mean G.P.A.s have increased. One factor, he said, is that faculty and administrators “want to make sure students do well” post-graduation, getting into top graduate schools and securing jobs of their choice. Particularly since the 1980s, “the idea that we’re going to grade more leniently so that our students will have a leg up has really seemed to take hold.”

Grades have also been pushed up by “pervasive use of teacher evaluation forms,” Rojstaczer said. “You can tell a professor that grading easy has no impact on their evaluations … and there are many arguments that say that’s the case, but the perception is that it does, so professors behave in a certain way,” giving higher grades to their students than they might if there were no evaluation forms. (This might prove especially true at institutions with high proportions of adjuncts, who are particularly vulnerable to losing teaching assignments if they don't receive high student evaluations.)

Another possible reason: students’ expectations. At private institutions, students are consumers expecting that their diplomas and transcripts be worth what they (or their parents) have paid for them. At more selective institutions, students enter with ever-higher high school G.P.A.s and “you don’t want the student to come to your office in tears for a B or C,” Rojstaczer said.

In their analysis of contemporary grading data, he and Healy found that, on a 4.0 scale, G.P.A.s at private colleges and universities were 0.1 point higher than at publics admitting students with identical combined math and verbal SAT scores. Among institutions with equal selectivity – measured by the average of the percentage of students with high school G.P.A.s above 3.75, the percentage of students who graduated in the top decile of their high school class and the percentage of applicants rejected – students at privates had G.P.A.s 0.2 higher than their peers at publics.

The data also support the commonly-held opinion that engineers’ G.P.A.s tend to be lower than those of students who major in the humanities or social sciences.

But the study does not take into account economic factors or broader national data, which is problematic to Clifford Adelman, a senior associate at the Institution for Higher Education Policy, who in the past has been critical of GradeInflation.com.

Adelman authored a chapter in 2008’s Grade Inflation: Academic Standards in Higher Education in which he argued that longitudinal data from the Department of Education’s National Center for Education Statistics suggested that grade inflation was not a major trend of the last few decades. “Unobtrusive national data are of no interest to folks who labor to build what are essentially quantitative anecdotes into a preferred story, and the unobtrusive national data tell a very different story.”

Rojstaczer and Healy’s study, he added, “doesn’t cite anything that doesn’t support a position based on fragmentary, fugitive data … and (with the exception of one article) completely ignores the economic literature."