In my last couple of posts, I have focused on the challenges of addressing grade inflation. To briefly summarize: given current data, it is hard to prove that grades are being inflated as opposed to simply compressed (i.e., merely a function of rising student quality). On the other, grades between departments are strikingly unequal, which presents a fundamental question of fairness. Why should students in Arts & Humanities or Languages be entitled to higher grades, on average, than students in the Sciences or Social Sciences?
After getting some feedback here and on Facebook, and after a good bit of pondering on my own, it seems that fixing the issue boils down to three options: force departments to meet some median requirements; require students to take more classes out of their disciplines; or change policies that make the interdepartmental differences less harmful.
The first of these departs from the principle of academic freedom. In the principal-agent relationship between the College and the faculty, the trustees delegate wide latitude to professors to distribute grades as they see fit. So long as faculty members are not engaging in untenable behavior -- discrimination by sex or race, or something equally egregious -- the College tends not to tamper with the grading policies of individual professors. Imposing required class medians would violate that relationship.
The first also imposes fairness quandaries of its own. An across-the-board median grade requirement could force professors to distribute grades lower than deserved. In the case of upper-level courses with specialized focuses, for instance, all students may (through self-selection) be quite apt and thus produce high-quality work deserving of high grades. Required medians may help distinguish the very best from the very good, but it would potentially devalue the very good into looking, on paper at least, like average work.
Requiring students to take more courses out of their departments could work. It stands to reason, at least, that the students of higher ability and effort (which GPA theoretically shows) will excel across disciplines. Those who have higher GPAs because they reside in high-median departments, par contre, would likely not fare as well in classes with lower median grades.
This is a bandage fix with its own externalities. This could narrow, but not eliminate, the gap since most students will still take more classes in their departments. And at any rate, students could probably find ways around the requirements by finding the highest-median options to meet the 'distribs' as they already do. This solution also raises serious pedagogical questions: it seems great if your goal is to expose students to as many disciplines as possible, but not-so-great if you want to accord students the maximum autonomy in constructing their own course of study.
Finally, the administration could change the policies vis-à-vis college honors. This, though also a bandage fix, seems the most readily implemented. The top X percent from each major would receive honors, instead the top students from the graduating class as a whole. This option doesn't fix the problem, but steals away its most potent venom.
Here's the big problem with all three fixes: they do not get to the root of the issue. Part of the reason for that resides in the institution-centric focus. Because of the academic freedom wrinkle, the College itself has little role to play in how different departments delegate grades. Real solutions can only be found after we establish why some faculty members dole out grade points more generously than others.
This is why I relapse into discussing grade inflation in economic terms. The College creates the currency (grade points), sets up general rules for their disbursement (e.g., that they range from 0 to 4 per course, that they can only be given in increments of 1/3, et cetera), and then delegates the power of distributing grade points to individual members of the faculty.
The question, then, is why faculty in some disciplines hand out these grade points more freely than those in other fields. If the work is not of higher quality, the logical conclusion is that grade points hold greater values in some corners of the college than in others; in other words, unless the quality of average language work is higher than average chemistry work, one must conclude that grade points are just worth less in the language departments ceteris paribus.
Why might that be? One could argue that the binary right-or-wrong nature of science and mathematics leaves less room for partial-credit, don't-worry-you're-getting-there grading. One could also posit that the difference in departmental culture produces different expectations of what "average" students should be capable of. Or maybe different departments have divergent ideas on what grades are for, and thus differ in how they distribute them.