Part I. For Information

The Senate Committee on Educational Policy appointed the Subcommittee on Grading during 1996-97 to investigate whether or not grade inflation had occurred and, if so, what steps should be taken as a result. The Subcommittee was unable to begin its work during 1996-97, but met a number of times during the later part of the 1997-98 academic year, gathered data on grade inflation from studies going back to 1955, and deliberated at length about what recommendations would best serve the University and its students.

Although conclusive data and research are difficult to come by, the Subcommittee is concerned that the dual factors of IMG (leading departments and faculty to seek to attract as many students as possible to courses) and reliance on teaching evaluations in promotion and tenure discussions could lead to increased pressures on faculty members to inflate grades. The Subcommittee recommends that SCEP carefully monitor the situation for the next several years.

The Subcommittee presents data with this report: graphs of average grade points by college over the last 35 years, including by course level for recent years and also including the University-wide trend (the thinner black line running from 1963 to 1997) (Appendix A), Grade Point Averages and Proportion of As by Designator and Level and average grades and percent A’s awarded by designator and course level (Appendix B), and grade point averages of graduates for the period 1987-1995 (Appendix C).

The Subcommittee also examined data on the characteristics of incoming students; information about them is attached (Appendix D).

Finally, the Subcommittee has noted a study recently issued entitled Unraveling the Complexity of The Increase in College Grades From the Mid-1980s to the Mid 1990s. The authors of the study, national in scope and based on student self-reports, concluded that "the average grades . . . rose from 3.09 in the mid-1980s to 3.35 in the mid-1990s" and that "at the same time, there was a drop in the number of hours students devoted to school work." They reported that the beneficiaries of higher grades tended to be Whites and females, and that while the grades at research universities were lower than those of students at doctoral universities in the 1980s, they were higher in the 1990s. Student effort was positively related to grades awarded. The increase in grades at research universities was only partly due to "grade inflation," however; more was due to changes in institutional grading structures. The authors conclude also that "students in the 1990s who devoted greater quantity and quality of effort were not systematically disadvantaged by receiving the same or lower grades than students who put forth less effort. This suggests that grades still reliably distinguish among students in terms of academic effort and performance. Moreover, grade inflation per se has not . . . reached the level where grading practices have devalued student effort."¹

The Subcommittee makes the following observations about the data it assembled.

1. Fairly dramatic "grade inflation" occurred in the late 1960s and early 1970s in almost all units. Following that period, grades appear to have remained relatively stable, although markedly higher than before the period of inflation. Since the late 1980s/early 1990s, grades appear to have been rising in some colleges and fluctuating in others. Table 1 reports the results of a simple regression of GPA against a trend variable for the period 1985 through 1997. Results for all but three units showed a coefficient on the trend variable that was greater than zero. A large coefficient is indicative of a strong trend in GPA increases. Ten of the 15 coefficients were statistically significant at the .05 level, and 8 were significant at the .01 level. The Subcommittee notes that there may be legitimate reasons for the increases in some colleges. Colleges with trends of increases of 0.02 grade points per year or greater included the Carlson School of Management, General College, the School of Nursing, Dental Hygiene, the College of Liberal Arts, and the College of Education and Human Development. (For comparison, the trend from 1963 to 1970 for the campus as a whole was an increase of 0.0556 grade points per year.)

2. The grade point averages of graduating students appear to be on the rise. In the case of the four Twin Cities colleges with the largest number of graduates, all are graduating larger numbers of students with higher GPAs.

3. Recent data on incoming students show encouraging trends (Appendix D). The "Average ACT Composite Score" appears in general to have increased since 1992. The average high school rank for incoming students has dipped and then risen over the last 20 years, and risen since the 1980s. The percentage of Twin Cities new high school students in the top quartile of their high school class has fluctuated.

Other considerations could come into play. It appears that grades began to rise before the apparent increase in the academic preparedness of students. Some also wonder if University class requirements are being diluted, if students are working less at their class work, or if high schools are demanding less. In the last case, characteristics of incoming students would be unrelated to rising grades.

To the extent that course grades are based on performance, and to the extent that the quality of students at the University has been increasing since the imposition of higher preparation standards in the late 1980s, then we would expect that average grades would be increasing somewhat. Not only did the University increase preparation standards, in the mid-1990s it also increased entrance standards, so that incoming first-year students are expected to have better credentials from high school.

To the extent that course grades are norm-referenced, we would expect to see no or very little increase in the average grade awarded (because the "competition" for grades on a scale remains, no matter the credentials of the group), and that there would be only random annual fluctuations around a fairly steady mean.

Subcommittee members had varied views on whether inflation of grades is occurring at present. There is no doubt in our minds that it occurred 25-30 years ago, and there was no subsequent deflation; grades rose and stayed high. The question is whether they are now going even higher. The graphs depict considerable variation: in some colleges, there appears to be a noticeable trend upwards in grades, in others there is no such trend, and in others the direction of grades appears to vary with course level (including some downward trends). In some instances, it appears that the vast majority of students in the field are close to achieving academic perfection. In our view, the judgment about the meaning of the grades, and the trends, must be made by the faculty in the departments and colleges.

The Subcommittee is firmly convinced that faculties in the departments and colleges need to consider what the grading standards mean for their courses and programs and majors. The wide variations in the average grades awarded by the different disciplines makes it clear that either standards and expectations are quite different or, we suspect, little or no thought has been given by the collective faculty within fields or departments to what grades should mean.

In order to promote what we believe to be the appropriate judgment and discussion, both within the departments and across the University community, the Subcommittee makes the following recommendations:

1. The Uniform Grading and Transcript Policy shall be amended to provide that every course syllabus shall include the grading standards set forth in the Uniform Grading Policy adopted by the University Senate (Appendix E) and the Senate policy on amount of academic work expected per credit (Appendix F).

2. Each college, department and program should discuss what the standards of the Uniform Grading Policy mean for its courses and programs, and what expectations the faculty have of students in their field of knowledge in order to achieve those standards. The Subcommittee believes that having such discussions would be a service especially for new faculty.

3. Data on the mean grade point average by designator and course level, on the percentage of As awarded by course level, and overall collegiate grade point averages should be prepared for grades awarded each Fall Semester. Data should be reported for all undergraduate students. Such a report can be seen in Appendix B.
   • Cells in the tables with fewer than 10 grades should be suppressed, in order to protect the privacy of students, but the numbers should be included in the totals.

   • The Office of Institutional Research and Reporting should see that required tables are produced each year and provide them to the chair of the Senate Committee on Educational Policy and to the office of the Executive Vice President and Provost.

4. The data tables and graphs required in (3) should be reported annually to the Senate. These data should also be provided to all deans and department heads, placed on a web site which is publicized.

The other way to think about grades is as criterion-referenced: there is an absolute standard of achievement established for the student to receive a particular grade. Thus, if the maximum test score is 100 points, the instructor may say that any score above 90 will receive an A - even though, when the test results are known, no student received an A.

The Subcommittee does not presume to know which of these two approaches to grading most faculty use. We suspect that many faculty may use an ad hoc compromise between the two approaches, so the grades in a course are partly norm-referenced and partly criterion-referenced.