Committee on Academic Affairs

Members of the Committee:

Please consider this proposal for implementing plus/minus grades. The current system requires that faculty discard considerable information regarding student performance. The system proposed here will result in transcripts that more accurately reflect what our students have achieved.

We propose the following system:

Letter	Quality
grade	points
A	4.0
A-	3.667
B+	3.333
B	3.0
B-	2.667
C+	2.333
C	2.0
C-	1.667
D+	1.333
D	1.0
D-	0.667
F	0

The design goals were:

• Increased resolution, i.e., more grading steps.
• Retention of the meaning of the current letter grades A, B, C, D, and F.
• Retention of the 4.0 scale.
• Least possible effect on grade point averages.
• A familiar grading system that offers the least change from what we now have while permitting more accurate representation of student performance.

The system we propose achieves higher resolution, with twelve steps instead of the current five. The letter grades without pluses or minuses carry the same value as in the current system, so the faculty who do not choose to use the new system many continue to grade as they always have.

"Plus" or "minus" adds or subtracts a third of a quality point to a grade. This system evenly spaces the plus and minus values between existing letter grades. Even spacing offers the greatest resolution.

Two grades are conspicuously absent: A+ and F+. We have chosen to omit A+ in order to retain 4.0 as the highest possible grade point average (GPA), while simultaneously retaining the meaning of A as 4.0 quality points. F+ was omitted, because we feel that failure is failure, and should carry no quality points.

We have conducted simple computer models of the effect of adopting this grading system on grade point averages. While the models adopt simplistic assumptions, the results are rather insensitive to the details of these assumptions. Therefore, we may have some confidence in these results. The models predict what one would expect:

• The GPA of nearly all students will be neither hurt nor helped by this system. Boosts in GPA due to "plus" grades well balance the depression of GPA due to "minus" grades. For example, a student who currently has a B+ average (3.333) may have achieved this by making mostly B's (3.0), while making A's (4.0) perhaps a third of the time. Under the new system, this student would likely make mostly B+'s (3.667) combined with an equal number of B's (3.0) and A-'s (3.667), once again achieving a B+ average (3.333).
• The GPA's of individual students will be more accurate representations of their knowledge and performance. All faculty acknowledge inaccuracy in the grades they assign. Rounding these to the nearest letter grade increases the average inaccuracy of the quality points assigned. Adopting a higher resolution scale lessens the inaccuracies introduced by rounding. The improvement in accuracy will be more substantial for semester GPA's than cumulative GPA's, since averaging over multiple semesters improves accuracy.
• Students with averages around a D- under the current system would have achieved a slightly lower GPA under the proposed plan. They would now occasionally receive a D- instead of a D, but will not receive the compensating benefit from the omitted grade of F+. This effect disappears for GPA's above 1.0.
• Student's whose current GPA is very close to 4.0 currently would have achieved a GPA as much as a few hundredths of a point lower under the proposal. There will be fewer students with a 4.0 GPA. While these students currently have all A's, a small fraction of these might have been A-'s, and will be awarded fewer quality points under the new system. (Including an A+ of 4.333 in the system would eliminate this effect, but also would allow some students to achieve a GPA of more than 4.0.)

Note that the above results ignore any effect of the change in grading system on the propensity of faculty to "round up" borderline scores to the next grade. Some have expressed concern that adopting plus/minus grades will result in grade inflation, since three times as many students will be in a borderline situation. There should be no cause for such concern:

• If faculty continue to "round up" all borderline grades in the same way, there will be no effect of overall GPA. Although nearly three times as many students will benefit by this boost each semester, the boost will be only one-third as large (0.333 instead of 1.0 quality points), so these factors will cancel.
• Faculty may be less inclined to round scores up to the next grade. Currently many faculty recognize the inherent inaccuracies in their grading and the large difference in performance between, for example, B+ and B-. Some choose to give students the benefit of the doubt if they have a score near the cutoff for the next letter grade. Under the new system, there is much less reason to round up. We have a name for "almost an A." It is called B+, with an appropriate quality point value.

Other possible benefits of the proposed system are reduced stress for some students and higher motivation for others. There is a big difference between steps in the current system, so the consequences of minor differences in performance can be substantial. Under the proposed system, minor differences in performance result in minor differences in grade assigned. At other times students find themselves late in the semester firmly placed in the middle of a letter grade range, so that performance on the final exam is not likely to have any effect on the final course grade. Under the proposed system, these students will find a reason to continue to prepare for final exams in the hope of making small improvements in their course grades.

We believe that there are no valid reasons to continue the current system now that grade point averages are no longer calculated by hand. The system we propose will better represent the performance of each student in each course. Faculty who do not wish to use plus/minus grades need not do so, and the grades assigned by these faculty will be perfectly compatible with those of faculty who wish to take advantage of the increased resolution. We ask that the Committee on Academic Affairs recommend to the faculty adoption of this system which will allow us to make these finer distinctions in the grades we assign our students.

Sincerely,

Page Laughlin

Dept. of Art

Rick Matthews

Dept. of Physics

Claudia Thomas,

Associate Dean of the College