I think data comparing grade inflation by discipline would also be interesting. For example, I would guess that in fields in which grading can be more objectiv

I think data comparing grade inflation by discipline would also be interesting. For example, I would guess that in fields in which grading can be more objective (consider the question of whether a computer program performs to spec vs. grading an analysis of some kind of literature), consistency in grading would be easier to maintain. Then there are considerations of what grading is for. If we consider grades to be certification that a student meets the standards of a profession, then majors leading to professions with lax standards might give higher grades.