End Grade Inflation - The Case at Chapel Hill

In February, 2000 the Educational Policy Committee at the University of North Carolina at Chapel Hill submitted a report, subsequently known as the Turchi Report, to the Faculty Council which then passed the resolution. The following are excerpts from this report. Quotations are given in regular type. Personal comments are in bold type. Some rearranging was necessary to produce a shorter but still coherent text.

The report begins:

The following exchange may sum up the state of grading and student grade expectations at UNC as we enter the new century. A faculty colleague recounts his meeting a former student who was manning the cash register at Foster's Market.

I took your course last year and it was the worst experience of my life.


Well, I mean, I enjoyed the course and I learned a lot, but it just about destroyed my GPA.

     (Fearing the Worst) What grade did you receive?

A B+..

On a more serious note:

The Educational Policy Committee was asked to examine the routine posting of the Carolina Course Review (CCR) on the World Wide Web for all to see. The CCR report showed that the mean expected grade as reported by students in their Spring 1997 CCR responses was 3.18 (not counting +/- grades)...

We were somewhat surprised by the high level of student grade expectations and consulted official University records where we found that student anticipations were not unfounded. Indeed, tracing semester grades back in time, we found that UNC Chapel Hill is in the midst of its second major grade inflation of the past three decades.

The following graphs are downloaded from this report:

In reiterating: The CCR report showed that the mean expected grade was 3.18 which implied an instructor approval score of 69.91 percentile points. If the expected grade were reduced by one standard deviation (0.415 points) the approval rating of the instructor would drop to 54.83, all other determinants of approval rating held constant. On the other hand, courses in which the expected grade average was one standard deviation higher led, ceteris paribus, to an approval rating of 87.2. In other words instructors who awarded an average grade of roughly 2.7 suffered, holding all else constant, a 32.37 percentile point disadvantage in student approval compared to those who awarded an average grade of 3.6.

Student evaluation of teaching (SET) scores are based on the class average GPA. Lower GPAs result in lower SET scores. The onset of grade inflation is then linked to the onset of SET usage.

In addition, the second bout of grade inflation began in the late 1980s, at almost precisely the time that the University mandated student course evaluations for all faculty.

The committee suggests: All student evaluations of instructors should be adjusted to purge instructor ratings of factors that are known to affect student evaluations but are not germane to assessment of the instructor's performance.

In discussing how to gain control over the situation the committee presents the following information, downloaded from the web site.

The horizontal scale value shows the GPA in 1987, and the vertical dotted line represents a sort of ideal grade average (2.7) that will be discussed below. The vertical scale represents GPA values in 1999 and the horizontal dotted line again represents the ideal GPA of 2.7.

The change which occurred in the distribution of letter grades between 1987 and 1999 is shown below.

Continuing: Grade inflation raises issues of horizontal and generational inequity. The grades of students who left UNC as little as a decade ago are no longer comparable to those of today's undergraduates. Perhaps more serious is the horizontal inequity across disciplines: Students who take a preponderance of natural science and mathematics courses will routinely receive grades up to a full point lower than their colleagues in the humanities.

Grade inflation erodes away at academic standards and traditional values.

Honorary societies are finding it increasingly difficult to use grade averages in identifying outstanding scholars. Phi Beta Kappa has had to raise overall eligibility standards three times during the 1990s, to the point that juniors will be required to have a 3.85 GPA and seniors a 3.75 GPA by August, 2002. See graph downloaded below.

Grade inflation makes it easier for students to attain a degree with a minimum amount of effort. While grade inflation makes it difficult to distinguish outstanding performance from the merely good, it also makes it difficult to penalize students who do little or nothing. If the standard grade for acceptable performance is a B then what does a C mean? In many cases, we suspect, students today receive a C for work that would have received a D a generation ago.

Kuh's analysis indicated that in all types of colleges and universities students of the 1990s reported spending less time on learning-related activities such as attending class and studying than did their predecessors but reported higher academic grades.

Since the minimum requirement for graduation is still a 2.0 GPA, grade inflation makes it much easier to have graduated without even contributing a gentleman's amount of effort. Lax grading ill prepares our students for the harsh competition that they will face after graduation.

Market forces operate on the individual faculty level. Use of SET drives this.

Instead, we have a classic example of what economists call market failure - the lack of congruence between the costs and benefits to an individual instructor of his or her grading policy and the costs and benefits to the University as a whole. The instructor can achieve private objectives by awarding high grades that subvert the grading system for the rest of the faculty and University. Moreover, at present the University has no mechanism in place to forestall this kind of behavior and to maintain the integrity of the system.

Market forces also operate on the institution wide level.

We do not suggest any kind of officially mandated curve; rather, we note that a University-wide undergraduate GPA of 2.6 to 2.7 would lead to a distribution of letter grades that, while not being unduly punitive, would better reflect the substantive meaning of the letter grades presented above. The 2.6 to 2.7 GPA range is consistent with research undertaken some years ago in the College of Arts and Sciences, which suggested that University-wide GPAs lower than this range would seriously affect student eligibility and progress to an undesirable extent.

The flood of As from the private colleges could well have an impact on our graduates' success in gaining graduate school admission or prestigious jobs. We will never be able to win a grade escalation war with our brethren in private colleges and universities, who clearly must be using high grades as a partial justification for their students paying tens of thousands of dollars for an education available at UNC Chapel Hill for a fraction of the cost.

The meaning of letter grades is defined:

Reiterating long-standing faculty policy we assert that the purpose of grades is to identify degrees of mastery of subject matter. Moreover, letter grades have specific meaning with respect to the mastery of that content.

�A�: Outstanding mastery of course material.

�B�: Superior mastery of course material.

�C�: Satisfactory mastery of course material.

�D�: Mastery of course material that is unsatisfactory or poor along one or more dimensions.

�F�: Unsatisfactory mastery of course material.

No mention is made of a grade scale that uses strict numerical cutoff levels such as 90%, 80%, 70%, 60% to determine letter grades. To continue:

We emphasize that we have not changed the meaning of the grading system. The verbal descriptions above are essentially the same as those that the faculty have, in theory at least, been using the past twenty-four years.

Grades measure performance, not innate ability or individual worth.

The report recommends that different schools within the university be held to the same standard, including those that have a minimum entry level GPA already in place, for example the Business School. The committee reaffirms its position on the meaning of the grade scale.

It is to distinguish degrees of mastery.

A recommendation is made that clarification of the new grading policy be printed on the face of the transcript.

The University of North Carolina at Chapel Hill strictly monitors its grading system in order to insure fairness and consistency both across units and over time. Therefore, the grades on this transcript reflect an overall grade average of 2.6-2.7. Special should be taken in comparing grades on this transcript with grades from colleges and universities that have not controlled grade inflation. See the distribution of grades on the back of this transcript.

Issues of concern are addressed:

This set of policies represents a comprehensive approach to grade inflation at UNC. It makes clear the meaning of each letter grade, both to students and their parents and to instructional staff. It sets clear quantitative guidelines that will achieve equity and fairness across instructional units. It provides for the regular dissemination of information regarding grading standards so that all can witness how successfully the common obligations of equitable grading are being met, it provides a mechanism to familiarize new and continuing teachers with university-wide grading norms.

At the same time, this program deals with some of the underlying causes of grade inflation. It provides a mechanism whereby student evaluation of teaching is divorced from factors that should not influence it. It provides a normative framework that will withstand pressures, from whatever source, to inflate the grading system. It addresses the issue of inter-university comparability of grading standards by using the student transcript as a "bully pulpit" to proclaim our commitment to integrity in grading. Finally, a program such as this would make it abundantly clear that this high quality institution recognizes its obligation honestly and clearly to report its qualitative standards to the world at large.

The report contains a reference to the importance of competition.

Do grades foster competition? Obviously, many people have goals that only comparatively few can attain; not everyone can succeed as a surgeon, a movie star, or a professional basketball player. Thus competition arises. And, surprising as it may seem to some, its effects are often beneficial. Gilbert Highet (referring to a boys' preparatory school) observed,

It is sad, sometimes, to see a potentially brilliant pupil slouching through his work, sulky and willful, wasting his time and thought on trifles, because he has no real equals in his own class; and it is heartening to see how quickly, when a rival is transferred from another section or enters from another school, the first boy will find a fierce joy in learning and a real purpose in life.

The Initiative to End Grade Inflation believes that competition can be better fostered without grades than with them. Top of the class is a noble goal. Attainment takes real effort. Staying there is downright challenging. Only real numbers, not letter grades, foster this level of competition.

In April, 2004 the Educational Policy Committee released a second report.

The proposal to set the mean GPA for all departments at 2.7 did not receive serious consideration and was not implemented. The report finds the increase in the GPA at UNC Chapel Hill similar to the national rate of .0146 per year. Faculty remain divided on the importance of the grade inflation issue.

A second issue is then discussed.

Since the 2000 report "a for-profit web service, Pick-a-Prof has begun to provide information about grade distributions in individual sections, listed by instructor, for classes taught at UNC Chapel Hill and many other universities. (Pick-a-Prof also provides a forum for students to post evaluations of courses and instructors.) Pick-a-Prof obtains grade information from the University Registrar under the State of North Carolina's public information law. At least one inquiry has been made to the Registrat from another organization about how to obtain this information in order to offer a similar service."

Students follow high GPAs.

from fall of 1998 to spring of 1999, a group (of researchers) at Duke University conducted a study in which information about grade distributions was made available to Duke students through a web site that tracked how students' examination of grades was related to their subsequent enrollment choices. The study concluded that grade levels had a significant effect on students' enrollment choices, with students being inclined to select courses in which the grades were higher.

Easily accessible information about grade distributions could also erode public confidence in whether the University is effectively fulfilling its obligations with respect to evaluating student performance. In recent years, some of the nation's most prestigious private universities have been subjected to ridicule in the press over the proportion of their students who graduate with honors. It is easy to imagine that information about grade distributions (at least for some classes or departments) could lead to the same kind of negative public reaction to grading at Carolina. For a state institution, loss of public confidence about the performance of a basic function, like evaluation of student performance, could undermine public support.

Public confidence is declining.

"For many students, college is a smorgasbord of easy courses chosen for their lack of academic rigor. there is no serious 'core curriculum.' Students spend limited time studying. Faculty and administrators make matters worse by allowing students to fill up their time with courses like UNC Chapel Hill's "Dogs and People: From Prehistory to the Urbanized Future' and 'Music in Motion: American Popular Music and Dance.' When students can get a minor in 'Social and Economic Justice' without ever taking a course in the economics department, it's hardly surprising that businesses aren't lining up to hire them." Jane S. Shaw, Higher Learning, Meet Lower Job Prospects, The Wall Street Journal, February 5th, 2013 - page A13.

The 2004 EPC report proposes a ranking system.

"A simple system might convert grades for a class into ranks indicating the percentage of students who did more poorly than a given grade in relation to the percentage who did better than that grade. A more sophisticated system would take into account the abilities of the students in a class, by weighing their performance in other classes, before grades are converted to ranks in the class."

A Quest to Explain What Grades Really Mean, Tamar Lewin, The New Your Times, December 25th, 2010. "A grade should mean outstanding work; it should not be the default grade, 'Mr. Perrin (a Sociologist at UNC Chapel Hill) said. 'If everyone gets an A for adequate completion of tasks, it cripples our ability to recognize exemplary scholarship.' As part of the University's long effort to clarify what grades really mean, Mr. Perrin (actually Dr. Perrin) now leads a committee (the EPC?) that is working with the registrar on plans to add extra information - probably median grades, and perhaps more - to transcripts. In addition, they expect to post further statistics providing context online and give instructors data on how their grading compares with their colleagues. 'It's going to be modest and nowhere near enough to correct the problems,' Mr. Perrin said. 'But it's our judgement that it's the best we can do now.'"

In the span of ten years, from the 2000 Turchi report until 2010, essentially nothing had been accomplished toward correcting the problem of grade inflation or the disparities that exist in grade distributions between different academic disciplines at UNC Chapel Hill. Between 2010 and 2013 public trust has eroded to the point the the public views higher education as "a smorgasbord of easy classes chosen for their lack of academic rigor." Jane S. Shaw - 2013.

This situation is not unique to UNC Chapel Hill, it exists nationwide.

At U (University of Minnesota) concern grows that 'A' stands for average. Jenna Ross, Star Tribune, May 27th, 2012. "Prof says grade inflation devalues grades given in more rigorous programs."