E.G.

"Me! Hear! My foreign ear–the sounds of welcome near!"

Do away with the ABCDF grading system at the undergraduate level.

December 30, 2004

Ignore the harmful phenomenon of grade-inflation, which is pushing us toward a de facto pass-fail system. Even assuming the minimum level of performance required for an A were not steadily creeping downward, the ABCDF system is still bad.

In the present system, the student who has done the best work in class receives the same grade as do the (often quite large) mass of students who have done the bare minimum required for an A. Likewise, in the present system, the student who just barely dips into the F-range receives the same grade as do those students who never bothered to turn anything in. We should have a much more precise way to distinguish among different levels of performance than this.

I suggest simply assigning a number between 0 and 100 to each student. This allows for fine-grained distinctions among students’ performance in an obvious way. Maybe I just lack imagination, but I don’t see any good reasons not to do this.

If we insist on keeping the ABCDF system (as I suspect we will), I think there should at least be a way to distinguish between the top student in a class and the other students (or, in large classes, between the top two or three students in class and the others). One way to do this would be to allow the instructor to dispense a limited number of "T" (for "top") grades in each class (e.g., one T grade per 50 students per class). This would provide an incentive for excellent students to continue to exert themselves even if they are certain they will earn an A with minimal effort.

Posted in Uncategorized

8 responses to “Do away with the ABCDF grading system at the undergraduate level.”

taylor

December 31, 2004

In my country, not all universities have the same grading scheme. My university employs ABCDF, but with + and – as well (with the exception that D takes the place of C-). This works just fine as far as I’m concerned. While a 100-point scale may work for math, I feel it is far too fine for philosophy (my field). I know a B+ from a B, but the difference between 76 and 77 is quite meaningless, as far as I’m concerned. If you’re worried about grade inflation, don’t do it! I’ve just finished assigning final grades for a class of 90-plus students. Only one student got A+. Six got A. I sometimes have a student complain that she only got B+ on an exam or essay, when she “always” gets A grades. I tell her B+ is a very respectable grade in philosophy. It usually turns out she’s a psych major, where an A grade seems to be as common as dirt.

LikeLike

Reply
Egarwaen

December 31, 2004

There are several issues with this:
1) It assumes that student work is purely and totally quantifible. This is very probably not the case. I can think of a lot of students in my classes who’ve done good work, but how exactly can you determine which was the best? This is one reason why the ABCDF system is used – it blurs the lines a little and eliminates noise from the grade distribution.
2) There’s the fallacy that one student MUST do the best work in the class. I’ve taken classes where this has been built into the grading system, and while I generally wind up near the top, I’m always totally burned out after such a class. Why? Because it turns grades into a /competition/. You don’t just have to learn the material well, you have to learn it BETTER than all the other students in the class. This makes the grades even more meaningless than they already are. Sure, I got 95 when taking Foo Bar 502. But a 95 is going to mean something completely different in a class of total slackers and a class of compulsive overachievers. (Unless you fix these things to specified performance criteria, at which point you’ve not really solved the grade inflation problem, merely added more resolution tot he system.)
This gets even worse when a professor literally enforces distributions, as I’ve had one do. With these schemes, if every student does the best possible work, everyone fails. The more students do good work, the lower their marks become.
I think the key is to simply admit that grading systems are utterly and completely arbitrary, and that the testing framework we currently employ is pretty useless for testing skills learned or knowledge retained. All it can tell you is what people DON’T know – not what they do.

LikeLike

Reply
david

January 1, 2005

Thanks, Taylor and Egarwaen, for the comments.
I agree with Taylor that some subjects don’t require fine-grained distinctions among students’ performance, but this can be accommodated by the system I’m proposing. An instructor could simply adopt the policy of giving grades in ten-point increments. (For instance, a great philosophy paper would get 100 points; a good one would get 90 points; etc.) I think this point answers Egarwaen’s first concern. Regarding Egarwaen’s second concern: As long as instructors are given the option of giving out T grades, but are not required to do so, the system I’m suggesting does not assume someone must be the best in class. But surely there are times where where someone is clearly the best in class. Why not allow instructors to express this?

LikeLike

Reply
zwichenzug

January 1, 2005

I guess I’m a little unclear on why grade inflation is such a bad thing. My best guess is that the worry has to do with lost functionality. So, the theory goes, in the good old days before grade inflation we knew how to pick out the wheat from the chaff, but now grades don’t give us much information at all. Is that close?
The problem is that grades never told us much anyway. I went to a state college in a farm state. How does my A in Russian History compare to an A- from Oberlin? How does it compare to a B in Calculus?
My own view is that grades are important, but not because they allow anything like a meaningful ranking of students. Their importance comes from their utility as a tool for motivating students. But if that’s right then the desiderata for grade distribution are quite different, and grade inflation may be appropriate (as may be your suggested T grades).

LikeLike

Reply
Egarwaen

January 1, 2005

zwichenzug – that pretty much sums up my feelings. How many employers these days care about grades? I’ve not found a single one. Pretty much all they’re good for is applying for future academic stuff and, even there, they’re used more as a way to make sure the student’s been applying himself/herself consistently.

LikeLike

Reply
david

January 1, 2005

If it turns out that grades are unimportant, then obviously grade inflation is not especially harmful. But grades are important to many people. They’re important to people applying to grad and professional schools. I suspect they’re also important when you apply to certain jobs.
Assuming grades are important, grade inflation is harmful because it decreases the expressive potential of grades. As a greater proportion of any given class earns an A, an A means less and less. Eventually, anything less than an A will be unacceptable. This will be, in effect, a pass-fail system: You “pass” if you get an A, you don’t if you get anything else.

LikeLike

Reply
taylor

January 1, 2005

I think of grading exams or essays as akin to judging figure skating. A panel of judges (even assuming they’re honest) will assign slightly different marks for the same performance. That doesn’t mean the marking is arbitrary. Expert judges know a good double salchow (whatever that is) from a mediocre one. As a result of years of study and practice, I believe I’m a fairly good judge of students’ written performances. Even in baseball, where the strike zone is pretty clearly defined, home-plate umpires call balls and strikes according to their individual perceptions of the strike zone. This is accepted as long as each umpire is consistent in his calls. The “strike zone” in academic grading is much less clearly defined than in baseball; hence my analogy with figure skating.

LikeLike

Reply
cmba

February 16, 2005

whats does a b c d and f means in the report card??
=(

LikeLike

Reply