I recently gave a quiz to my evening MBA students that turned out to be harder than they expected. Actually, it was harder than I expected it to be too. So, I have been receiving a flurry of emails and concerns about it. To not think much about this would be callous, and there is no getting away from worry, even if it is someone elses.
That everyone found the quiz hard (as I know from grading it) means that relatively speaking, no one has actually done as badly as they think they have. In fact, since grading is relative, it is true to say that most have done better than they think they have.
The format of the quiz, since it did not follow closely the back of chapter problems, but rather, my presentation in class, might have thrown folks off. But, each and every question was specifically based on something I repeated more than once in class. I set these quizzes with a view to them being a summary of the most important concepts. In my mind, the role of a quiz is to provide a break point for study (consolidation and closure) and to give feedback on what students know and do not know.
Students work very hard, and I can see why it is disappointing to not crack a quiz after lots of work. In addition, being evening students means many have to deal with job idiosyncracies, making education a harder slog than it should be. So there is a great worry about getting grades below a B+. In fact the average cut off to graduate is a B.
We use scores to rank students (so its relative), and then decide cut off scores after calibrating for the level of difficulty entailed in the course. If the test is hard, it just moves the average, but not the ranking. And now for the statistics: unless a student knows close to nothing (a C grade or worse to me), we have the following range of grades to work with: A, A-, B+, B, B-. If we center the distribution on B+ (which is also the cut off GPA for meeting certain requirements) then 25% or more of the class will be below the required level. That clearly is untenable as it seems to me that MBA programs across the country target the percentage of students failing to meet requirements at less than 5% (maybe even less). So where do I need to center my distribution so as to make sure that across many classes only 5% of students get below B+ on average? Turns out it needs to be quite high, making a B- very far out in the left tail of the distribution. And therein lies the dirty secret of grade inflation.
Of course, it takes little to realize that grade inflation does not make students “feel” better – it only makes them feel worse, for two reasons: (a) it leaves them wondering whether they really deserve the high grades being handed out, and (b) it makes everyone think a B stands for Bad. When I went to business school I got loads of Bs and a C too. I ranked high in the program and my GPA was only 3.4! If you simply compared our grades to today, we were relatively a bunch of miserable underachievers! But we did not have to worry about grade inflation. I am not sure when these grade cut offs entered business schools, but they cannot be a good thing. They seem to be a way to correct for admission errors, but end up imposing harmful externalities on all students. Professional degrees are better off being managed on a pass/fail system. Last time you went to your doctor, did you ask for his/her GPA?