TABLE OF CONTENTS  |  FRONT PAGE  |  NEXT ARTICLE |  PREVIOUS ARTICLE  |  UNC HOMEPAGE

RELATED STORY


Faculty debate grade inflation

A Carolina professor runs into a former student who tells him, "I took your course and it was the worst experience of my life."

"Oh?" says the professor.

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

Fearing the worst, the professor asks, "What grade did you receive?"

"A B-plus."

It reads like a joke, but the punch line has serious implications according to Boone Turchi, the professor who used the story in a Faculty Council Educational Policy Committee report to highlight the problem of grade inflation on campus. Turchi presented the report to the full council on Feb. 11.

The report concludes that the University is in the midst of the second period of major grade inflation of the past three decades and that steps should be taken to curb it.

The average grade point average during the spring 1999 semester was 3.0. In 1987, the average was 2.7. Generations of Carolina undergraduates completed degrees with GPAs averaging 2.4.

Another troubling statistic is the changing distribution of letter grades, Turchi said. In 1987, for instance, A's accounted for 23.4 percent of the permanent letter grades students received; by 1998, the percentage of A's had climbed to 38. As the number of A's increased, the number of C's fell -- from 26 percent in 1987 to 17 percent in 1999.

Data shows that the highest grades came from professional schools, arts and science departments, humanities and social sciences. Professors in natural sciences give the lowest grades.

The report provoked a lengthy discussion during the Feb. 11 Faculty Council meeting and strong doubts about the veracity of its findings remain, Turchi knows.

Faculty members seem to disagree not only on the causes of grade inflation, but on the question of whether it is a problem to be solved or a sign of increasing success that should be celebrated. That's part of the problem, too, Turchi argues.

The Educational Policy Committee began its study in response to the posting of the now-defunct Carolina Course Review on the web. Committee members found that a student's rating of a professor's performance was strongly affected by the student's expected grade. At the same time, the second outbreak of grade inflation began in the late 1980s, precisely the time the University mandated student course evaluations for all faculty.

Ed Neal, the director of faculty development at the University's Center for Teaching and Learning, said in an interview after the Faculty Council meeting that he does not take issue with the facts in the report so much as does the conclusions that the committee draws from them.

Neal argues, for example, that the report overreaches when it suggests that posting the Carolina Course Review on the web contributed to grade inflation. One may have followed the other, Neal said, but finding a correlation is not the same as proving a causal effect.

At a research university, decisions about promotion, tenure and merit pay are more closely tied to the caliber and amount of research a professor generates. Having a reputation as an excellent teacher may be appreciated by students and respected by peers, but it is not the biggest carrot on the stick in academia, Neal said.

That may be true, Turchi said, but Neal may be overstating the insignificance of teaching. Departments will not promote people who do not pass the teaching test, Turchi said.

Neal said he accepts the report as a way to start further investigation. "I don't think it is the final word," he said.

It is on this one point that Neal and Turchi can agree.

Turchi said after the meeting that he did not expect faculty members to accept the report without question. But none of the challenges that he has heard so far has convinced him that the report is that far off the mark.

"Nothing was brought up (at the Faculty Council meeting) that we (committee members) hadn't thought about already," Turchi said. "Most of the issues were treated one way or another in the report -- maybe not to the satisfaction of everyone."

At the same time, Turchi said, there has not been a groundswell of negative reaction. Instead, many professors have told him that they have been worried about the same things for a long time.

One central concern that will need to be addressed is making sure Carolina students would not be penalized when their adjusted grades are judged against the inflated grades handed out at other universities. "We are going to be aggressively approaching that problem in any recommendations we make," Turchi said.

Ultimately, Turchi wants to see the Faculty Council play the lead role in developing guidelines to address the problem. He hopes a set of proposals can be presented to the council by the end of the academic year.

"We need to get the issues out more fully and more clearly," Turchi said. "We hope to have an open forum for students and the entire University community to talk more publicly about this."

Grading: Is it art or science?

Neal thinks that part of Turchi's mistake is that he expects grades to be more than they are. Grading is not a scientific experiment. What that means is that grading is incapable of producing the kind of exact measurements of student performance that Turchi expects it to, Neal said.

Neal sees grades the way Paul Dresser described them 25 years ago in an article titled "Grades: One More Tilt at the Windmill." Said Dresser: "A grade is an inadequate report of an inaccurate judgment by a biased and variable judge of the extent to which a student has achieved an undefined level of mastery of an unknown proportion of an indefinite material."

Neal said the committee bases its analysis on a normative grading scale that depends on the performance of one class as a whole during one semester. But the report fails to take into account the classes with a handful of students that would not represent statistically valid samples.

"When considering the University as a whole, you need to take a more multi-faceted look at the data," Neal said. "You can't just cram it all in the same pot and say it's grade inflation."

Neal said the report does not consider another grading model that is called the "criterion-referenced system" or "mastery learning model." Here, the professor sets specific objectives to be met and ties each student's grade to how well and completely he or she meets them.

Professional schools are designed more and more to make it possible for every student to master the requisite skills and knowledge, Neal said. Under such an arrangement, a course in which all students end up earning A's may be the result of nothing so terrible as a professor and students doing all that is required of them to do.

Turchi is convinced that Neal misunderstands what the report is saying and what it seeks to accomplish. Under current faculty guidelines, the grading system seeks to differentiate between levels of performance between students. Under a "mastery model," mastery would be worthy of a C, not an A. The C's would accurately depict all students as performing satisfactorily. And that should be OK, Turchi said.

"If mastery model leads to all students getting A's, the grading system as we now know it is being subverted and being misapplied," Turchi said.

Turchi said the faculty currently operates under a set of policy guidelines in which grades are supposed to be used to precisely evaluate degrees of mastery and performance between students.

Such a system attempts to distinguish those students who do outstanding work from the students whose work is merely good. The former should get an A, the latter a B. The same kind of sharp distinctions should be made all the way down the grading scale. And that does not appear to be happening as much as it should be, Turchi said. And it is happening in some departments less than others.

A "gentleman's C" used to go to students who at least put forth some effort, Turchi said. Today, the old gentleman's C is closer to a gift granted to students who do not try at all.

"When so many departments show the vast bulk of grades as a B or better, such distinctions are rendered almost meaningless," Turchi said.

Larry Rowan, a physics professor and director of the Center for Teaching and Learning, said the committee's report is based on parametric statistics that assume a normal distribution of results along equal intervals of measurement. The normal distribution, however, assumes there will be uncontrollable variables occurring at random. And in Carolina classrooms, that is not the case.

"This is a selective institution," Rowan said. Only high school students who are intelligent and motivated get into the classes.

Combine good teaching with bright, hard-working students and what you can expect to get is a cluster of grades at the top end of the scale. That's not grade inflation, Rowan said. It's the inevitable consequence of controlling two key ingredients for success from the start.

Turchi said that what Rowan is talking about is grading on a curve. "We are not advocating or asking for that," he said.

The goal, Turchi said, is for the system of grading to be consistent across departments. That is not the same as saying that every class should have an even distribution of grades every time.

But now, when each department applies its own standards to grading, inequities are created about what grades are supposed to signify. Turchi argues that an A in one department should count for the same kind of extraordinary achievement as it would in any other.

Raising the bar

Neal cited a host of factors that the education committee failed to consider that could be contributing to rising grades -- from students taking lighter course loads to the rising SAT scores of incoming freshmen to the rising percentage of women undergraduates.

Women now represent a little more than 60 percent of undergraduate students. Statistics show that women generally out-perform men at academic tasks, Neal said.

Turchi said he accepts the evidence that points to the conclusion that students are coming to the University better prepared than ever before. But, he said, "it's irrelevant to our arguments."

If students are continually getting better and faculty members are not keeping up with them by raising standards to challenge them, then the faculty members are failing students in a way that is more damaging than a bad grade, Turchi said.

The worst kind of grade, in Turchi's view, is the one that misleads a student about his or her real accomplishments in a class. And it diminishes the University in the process.

"A fair and honest grading system is a major step toward improving the quality of the University," Turchi said.

Report online

To see the Educational Policy Committee's full report on grade inflation, go to: http://www.unc.edu/faculty/faccoun/reports/R2000EPCGrdInfl.PDF


TABLE OF CONTENTS  |  FRONT PAGE  |  NEXT ARTICLE |  PREVIOUS ARTICLE  |  UNC HOMEPAGE