"If asked to predict the pattern of grade inflation, I would have predicted the opposite trend. In my own experience, graduate students tend to be the harshest critics of undergraduate work, folloed by junior faculty (tenure track or not), followed by senior faculty"

That was my experience as well. Some of the toughest instructors I had were the youngest. They were also often the ones who went out of their way to kick me in the ass, which is actually pretty cool. Nice to know somebody cares :)
The oldbies often had the go along get along attitude, which I also appreciated, particularly at the 8 AMers.
Course that was way back in the 90s. Im sure these kids today are taking everything to hell.

Education today ( and this was not always so ) is the place where people are supposed to develope the commitments of citizenship, the knowlege to function in a technological society, and the moral temperances to function in the complex and highly conceptual social continuum that is our nation. These crucial functions are boiled down to a GPA or an SAT. Complete contempt for such an arrangement is the surely a sign of sanity.

More simply, I belive it boils down to the pressure put on schools to graduate kids. If there are any reasons beyond academic that lead to a child being passed to the next grade, you're doing the child a disservice. But, anyway, if schools want to see more kids "do well" what better way is there than to inflate GPAs?

Well, in my experience most grad students start out as tougher graders, but then become softer for two reasons: (1) external pressure from faculty and students; (2) growing cynicism with the system, less incentive to grade according to some academic ideal, more swimming with the stream. I know that I lost all my respect for grades when I began grading.

Causes are likely to be complicated.
1) Prof. Drezner is at least partially correct. In the last decade, there has been a push to improve undergraduate teaching. As part of this, teaching performance has been given increasing weight in hiring/promotions decisions. To evaluate teaching quality, many universities incorporate surveys of students. Especially at large schools, this has elements of a popularity contest. Grading, even unconsciously, probably influences the popularity of instructors.
2) On the other hand, some of this may represent a real increase in student competence. I'm impressed with the number of very good students coming out of high schools, often with significantly more advanced preparation in the form of AP courses, etc. While a lot of high schools perform abysmally, a lot are doing better than ever.
It would interesting to see the variance in Thoma's numbers (not in his charts). If the rise is due to an increasing number of very competent students, then variance might be rising and the grade distribution should exhibit more skew.

grad hit this one out of the ballpark. Remember, Dan, most grad students/newbie teachers start off as teaching assts before we are fully in charge of a class. I rarely got support for taking a semi-hard line on an obviously cheating/slacking student. I remember one kid who hadn't showed up for a single discussion group (mandatory) turned in a mediocre midterm and was many hours late on his critical takehome final. I wanted to fail him, but the prof in charge inflated it to a D. Out of fairness, this bumped everyone else up a grade, which meant everyone was "above average." Disillusionment inevitably follows enough such instances.

Let's call it like it really is: many schools in this country have devolved into degree mills. Pay your money, collect your degree. For kids in the top schools, this is less of a problem. Here intense pressure from parents (sometimes direct, sometimes hinted at by the student to either a teacher, administrator or anyone who'll listen) to get unearned As so Sally can get into law school make it all but impossible to give Sally the B- she so richly deserves.

When I was a TA, the professor gave me the desired grade distribution and I hit the mark. My students would complain that I was a hard grader and I would say I hadn't picked the scale. What ticked me off is when students would complain to the professor, who got to play the magnanimous man to my shrivelled-nuts loser and raise the grade. Jackass.

It's really hard to get anything lower than a B at the graduate program in which I am enrolled. At one point a professor stretched the C+ margin to 24/180 (the lowest grade), with a 160 or above being an A+. Quite a distribution, that.

I also took some graduate courses while an undergraduate at another University, where among the graduate students there was a general consensus that anything below a B was the equivalent of failing. That probably stemmed from University guidelines that endangered financial aid for students below a 3.0 -- what department wants to find more funding for its mediocre students? Let them take loans....

Not that I, lazy as I am, really see anything wrong with it. There'll be plenty of time to sweat over performance -- when I'm getting paid to do so.

(Both institutions, for the curious, are second-tier research institutions. Top 25 or 50, depending on the department, with enough name recognition that my resume doesn't look completely empty.)

The differences Dan notes between senior and junior faculty also shows up in scholarly journal referee reports. In my experience as a managing editor for one journal and also as a sometime referee for others, junior faculty (and sometimes senior grad students) make more rigorous article referees and book reviewers than their senior colleagues, who often enough tend either to pull their punches or to scratch the backs of their colleagues.

In fact, grade inflation in the classroom may be far less of a problem than content inflation in scholarly journals. Some articles, of course, should never have been published at all. But the major problem is that too many referees, authors, and journal editors lack the time or patience (or sometimes the guts) to do a proper job on the referee and revision process.

I believe that grade inflation is real. However, I have noticed that many full professors, especially the "stars" of the profession, do not care about teaching. They present the material in lectures and the students are pretty much on their own. If they get it fine, if not, these professor's don't seem to mind. For many of these people, having students figure out what is expected of them is just another part of the evaluation. Hence, you get lower grades. Many younger faculty members are committed to being good teachers (even if they don't actually suceed). They often feel responsible for making sure that students learn the material. Thus, they spell out expectations, offer study aids, etc. Students do better. In short, I think there is a generational difference in approach to teaching with the new approach resulting in higher grades. Like all generational difference, this process is uneven and not universal, but it is real, and accounts for _some_ of the rise in grades. So students aren't necessarily better, but schools put more effort into helping students do well.

Any chance the difference found among the ranks is due to the grading standards when that cohort entered the professoriat? If grade inflation has gotten worse over time, perhaps those who are more senior are grading according to the standards they graded with when they started, where as the newbies are grading with the current, laxer standards.

But I do think that the role of teaching evaluations is central. I am willing to give tougher grades now that I have tenure (and found myself relaxing my standards when I went to a new institution and had to go through the tenure process again; I tightened them again after I made it through the process). My teaching evaluations are good but I still get complaints about my tough grading standards - which are much more lenient than they were at my previous institution just because of the local norms and because teaching evaluations still matter for promition.

I also have a friend who tried an experiment with one course he taught identically in two different semesters -- the second semester he raised the grades considerably and, sure enough, the student evaluation numbers went way up. If your institution takes these seriously you'd be foolish not to . . .

I'm not a teacher, but it seems like we're overlooking something here. When I was in school (during the 70's) it was much tougher to get an A in a science or math class than in sociology or English. Indeed, some of my friends who couldn' t pass physics to save their lives, eventually graduated with liberal arts degrees with 3.7 averages. You can't discuss grade inflation without also discussing what subjects the students are taking.

I agree with Joel that this also seems to show up in scholarly peer review. The reason is similar to what Dan said about grad students being harshest on undergrads; they've been learning how to be critical, but have not yet had to do an original project of their own so they do not realize that some compromises usually have to be made regarding research design, etc. That is, maybe with tons of money, a large group of research assistants and lots of time someone can pull off the ideal scenario of a project, but since those circumstances are rarely the case, most projects will have limitations. Grad students seem to be the least sympathetic to that.

I experienced this first hand when I presented findings from my dissertation at a workshop for grad students one summer - I was still a grad student, but close to finishing up. The students in the audience were super critical (not realizing that pretty much every critique they brought up I was completely conscious of) while the faculty present were very encouraging. It would be very interesting to go back and see what types of projects they have developed since.:)

It is also my experience that grad students are harsher graders than full professors. I suspect that there are many bitter graduate students out there all too eager to take out their frustrations on undergraduates. I wonder if it is their low pay and the bleak academic job market. Maybe they are angry at the full professors who hang on forever and won't make way for aspiring graduate students, so instead of going home to kick the dog, they shift their emotional component from their superiors to lowly undergrads. The psychological term for this is called "displacement".

I'm not sure the "degree mill" argument covers all the bases. You can get a degree earning C's, and many students are happy to do so.

The getting-into-grad-school argument makes more sense. Professional degrees are more and more in demand, and GPA has been a prime determinant of admission. So customers (students and parents) will demand higher grades under the assumption that this will help the little darling get into law school.

Grade inflation, though, has a short term impact. Top universities start giving out A's like Snickers at Halloween, and other universities quickly follow suit. Soon, graduate programs start getting applicants with 3.7 to 3.9 GPAs when they used to get 2.9 to 3.9 GPAs. They take this into account, and they start to find new ways to differentiate between applicants. This is unfortunate, because undergrad grades should be the primary factor in determining grad admissions.

The interesting thing about "grade inflation" is that it has an upper limit: 4.0 at most universities. (No one's applying to grad school with GPA of 3,000,000.5) So "grade inflation" has had the effect of compressing GPAs to a point they've become much less meaningful. It's really not "inflation," it's simply the elimination of grades as a meaningful differentiator.

Andrew Steele has a really good point. It isn't grade inflation it's grade compression. The Economist wrote an article that pointed that out a few years ago.

The other thing half the posters point out is that there a incentives for people to not grade in a discriminating way.

Universities provide two functions, training and evaluation. Perhaps the two should be split. Indeed the increasing importance of GREs, GMAT, LSATs and so on are doing exactly that.

But universities have an interst in this not happening, so perhaps they will start grading on a curve, as Australian Universities do. This also has problems, as Larry points out, grades are not equally easy to get in all subjects.

Oh, and I should have remembered to point to this related article by Jordan Ellenberg in Slate in 2002: Don't Worry About Grade Inflation published in his "Do the Math" column.