I teach computer science courses, many of which could be offered by a math department. Despite the religious devotion some of my colleagues have to partcular numbers (a 65% must be a C), I set the standards for letter grades on each assignment based on more-or-less objective standards:
1) Since a C or better allows entry into courses that use mine as a prerequisite, I have an operational definition of a C -- the minimal performance that lets me say to my colleague in the future course that I judged the student capable of going on.
2) I give an A for a performance that represents everything I could expect for a student with the normal background for the course. Lots of students exceed this standard because of unusual background ("ringers"), but my university doesn't allow A+ grades. We do have a system for recording "class distinctions" at the department level, for later use by recommenders.
I don't "grade on the curve" in the usual sense because I have no preconceptions about how the class will perform relative to these standards. I could in principle give all A's, or pass no one. (The latter would have repercussions, as it would obviously represent a failure on my part!) In a class of 30 or more, of course, the distribution of grades will have some effect on what I decide is a C or A performance, because of the feedback it gives me on how easy or hard a particular exam or homework set was.
I'm tired of curmudgeons who insist on archaic uses of particular words or phrases against popular practice
CRAZY TALK! THIS IS CRAZY TALK!
In a class of 30 or more, of course, the distribution of grades will have some effect on what I decide is a C or A performance, because of the feedback it gives me on how easy or hard a particular exam or homework set was.
I swear I am not picking on you, but we humanities types cannot get away with this shit. If I give an exam to thirty students and the average grade is a 50%, it means I designed a poor exam or taught poorly.
I suppose this has something to do with the idea that science classes have been standardized, and you can make sense of 'all the material in organic chemistry' in a way you can't of 'all the material in ethics', but it feels like science exams are designed with the entire knowledge of the discipline in mind, rather than what was taught in the course.
3: I pretty much agree. While some words or phrases have inescapably changed meaning, the best way to preserve currently existing meanings (and hence precision in language) is not to pick your battles, but rather to assume that all changes of meaning are a mistake and oppose them as such.
4: what about a course with 5 sections and 1200 or so students?
Your promotion and tenure committee actually looks at the grades you're giving? Wow, to think they care about teaching even that much. I'm genuinely surprised.
6: He's the one that said 30, not me. If you have 300 students per class, they're not learning anything from you anyway.
Grade inflation is a hard problem because there's individual incentive to inflate-- it changes evaluations and saves time, since no one complains about a high grade-- but if everyone does it we're screwed. I say let those with tenure make the bold stands.
At my current school, despite their usual hardcore-econ solutions for every problem, they took the regulatory way out on this particular problem. For the MBA classes, the highest average grade that a teacher can given their sections is a B+ (3.33 on our scales). In practice, this usually means that the highest person or two across all the sections gets an A+, the rest of the top fifth to a quarter of the sections get As or A-s, then a whole lot of Bs of different varieties and some Cs and Ds for particularly awful performers.
This at least makes the A worth something, and it's quite rare for anyone to have a 4.0 in the program. But at the same time, the huge number of Bs is necessary because businesses usually refuse to reimburse employees for any class where they get less than a B. If any business school started handing out lots of Cs, they'd have a huge student crisis in no time.
3, 5: Get with the times, oldsters. Labs is totally right.
9: I should also note that each section of these courses is typically 40-65 students, and that a professor will often teach 2 (and sometimes 3) such sections during their teaching term. So this also drastically reduces the likelihood of sudden really good classes or really bad classes, due to natural dampening of variability as a percent of the whole with larger N, etc. etc. And that's why a severe hardcap on our professors' grade averages doesn't end up as badly as it would if they only had 30-40 students at a time and there was a larger chance of getting 15 students deserving an A in one freak quarter versus only 4 deserving an A in another.
We will bury you, teo. [bangs shoe on podium]
I think the strategy of grading about the same as everyone else makes a lot of sense, especially in the humanities and social sciences.
Grades have two distinct roles. They are there to sort students on the basis of their accomplishments for outside evaluators and they are there to goad students to work as hard as they can to learn the material. The grade inflation controversy tends to overvalue the sorting functions of grading and undervalue the goading functions. Too much grade inflation removes all incentive to work hard to learn. Too stringent a grading scale leads students to avoid the class, so they don't learn the material at all. (There's really no excuse for any introductory class to have a high score below 70. If there is, either the teacher is incompetent or the expectations are unreasonable.) A middle of the road grading policy gives some incentive to the very talented to work hard to rise above the mass, while also giving some incentive to weaker students to push themselves to learn more without dread of failure.
Of course, one should as a teacher have some sense of what constitutes an acceptable level of accomplishment that defines the bottom of the grading scale.
The screening function shows up eventually, because screening is one of the main things education is about. In the humanities screening takes place during the grad school application process and grad school, especially the first year. That's why I'm skeptical of higher ed as the road to equality.
When low-ranking departments within a school upgrade one of the first things they do is flunk out a bunch of students.
A lot of faculty at my Last Chance Urban U. alma mater (not Reed College) saw flunking out students as a duty. They also looked for students to mentor, but they were definitely big on the gatekeeper role. (Others gave everyone useless As.)
They don't require philosophy at your school? That would solve all the world's problems, including the most serious one currently facing us: that this post will receive much blogostratic attention; but not a single person will attempt to address the statistical arguments on their merits.
People who want to preserve distinctions and precision by insisting on the traditional definitions often also seem offended by neologisms which express new distinctions. I think that conservativism is more the motive than desire for precision.
Damn. I read that post, and while its conclusions agree with my layperson's understanding of the issues, I just don't have the chops to actually agree or disagree with the substance. I'd guess that 99 out of 100 readers are in the same boat.
they are there to goad students to work as hard as they can to learn the material get good grades.
19: "grades" s/b "jobs"
Which ties back in to the "sorting mechanism" function, of course.
I disagree with Rob's assessment of undergraduate science teachers. They hate the gatekeeping function and resent the fact that it rules their teaching. If your institution is a pre-med factory, you may end up teaching to the MCAT -- that's annoying.
Tenured profs strike back by teaching what they want, when they want to. TAs strike back (here, it's not love, it's perverse pleasure) by trying to slam the gate closed. It's the passive aggressiveness engendered by knowing that a third of their students will be making much more money and be considered a real "doctor".
My law school had a very strict curve for first years. It was modified a bit my second year, because many of the junior faculty argued that its harshness was affecting students' abilities to get jobs.
One of my professors --who was emeritus--gave exclusiovely multiple choice exams and didn't bother with the curve, but his grades weren't that far off from the curve.
The curve for upper division courses was rarely as strict. My partnership tax class had only 5 people in it. We had a take-home exam, and we were allowed to talk to eachother about it. I think that it would have been pretty hard to curve a course like that. In fact, my professor said that he didn't even try to curve either his partnership or his corporate tax classes, since he figured that the people who chose to take upper-division tax classes were self-selecting and likely to be pretty good. Its also possible that if he'd been super harsh, his enrollment for partnership tax would have been zero. Corporate had more people in it--maybe 20.
Some of the natural sciences have a captive market of would-be doctors and so have no need to worry about enrollments
Though, when my undergrad institution was up for recert, they found that humanities gave the highest grade yet expressed the lowest satisfaction with their students, while sciences gave the lowest grades and expressed the highest satisfaction.
Actually, now that I think about it, I'm not sure that's inconsistent with your assertion.
Fraid I can't hang around, but your grading sounds a lot like what I do.
I was so busy looking at baseball players' cocks that I forgot to thank you for the shout out. Sorry.
Don't worry about it, rob. Happens to the best of us.
Also, I've found that the easiest way to get your grades to fall into a neat curve is to give a lot of assessments. Much of the seeming randomness in grades comes from essentially undersampling students' abilities. Essentially testing is a problem of random sampling: you are judging what they can do by looking at what they actually do, which is a small slice of what they can do. If you increase the sample size, you increase reliability. Student grades start to zoom in on what they actually know they can do, and people complain less.
27: That's similar to what one of my favorite professors did for a history-of-philosophy-type class. At the end of each class he assigned a question pertaining to the next chunk of reading, and we had to prepare a brief, typed answer (say, a paragraph) for the next class.
He collected them, read them, commented on them, and returned them. And each one counted for a few points, but over the course of the semester, those points add up. Seems like a lot of grading for a 30-person class, though.
All good teaching practices are very labor intensive.
27: right, the down side is grading them.
28: I do something like this, but I find it hard to grade them in a meaningful way rather than check that they've been done with at least a halfway serious effort.
21: Neither the tenured faculty policy of avoiding intro level teaching nor the grad student policy of slamming the door is an alternative to viewing intro level courses solely as gatekeeping courses. They are just responses to the assumption that the only thing you do in an intro level course is screen people out.
Neither your tenured faculty nor your graduate students are using intro level courses as an opportunity to teach the general public the importance of your field and its role in society and people's lives. Philosophers have to do this every day. Interestingly, I also spend a lot of time doing this for the hard sciences, because they don't do it themselves, and when they try to they do it badly.
they don't do it themselves, and when they try to they do it badly
Sheila Tobias writes and talks about this. Prestige in the sciences comes from having successful in-field students, is part of the problem.
31: Is part of the problem maybe that for physical sciences, the classroom exercises have very little to do with actual science? You learn more about chemistry (my degree) by doing actual lab work than in mechanism problems, or learning nomenclature. And that's also where a lot of the teaching takes place, especially with young profs. They work in the lab with their grad students and undergrads. They teach but in group meetings and at lab blackboards.
Additionally, intro chemistry classes have more in common with intro foreign language classes than intro philosophy classes. You're learning basic vocabulary and grammar. You aren't learning any higher order concepts or big ideas or repeating any important experiments.
The gatekeeping in (non-GenEd) STEM courses is not so much screening out undesirables. Most of our courses build on one another. There's a sequence of courses. The teacher in the later courses must make assumptions about how much the students in her class understand of the material that was covered in earlier classes in the sequence. If I pass on to her a student who is clearly underprepared, I'm setting her up for failure. If I pass on to her a student who is only marginally prepared but thinks he's hot shit, I'm similarly setting her up. This is not collegial behaviour.
It's not the pre-meds I'm worried about. It's the regular majors. (Though I did once teach a Calc I that was majority pre-med; I used to think I shouldn't enter the room without a chair in one hand and a whip in the other: "Back, back, I say!")
19: "they are there to goad students to work as hard as they can to learn the material get good grades."
True. But that's exactly where things come back to the professor, and their ability to make "getting a good grade" roughly synomous with "learning the material".
I suppose it may be the disconnect between those two things in a lot of classes that makes grades get a bad rap.
30: "I do something like this, but I find it hard to grade them in a meaningful way rather than check that they've been done with at least a halfway serious effort. "
Bingo. Same boat, exactly.
Though I've started to think that this is actually an okay compromise. When you have a lot assignments, you don't have to agonize over the details the way you might if the whole grade hinged on how carefully you graded 2 exams.
To the extent that profs don't assign work because they don't want to have to grade it, it's probably better to just assign it and then use the "halfway serious" method.
I'm just a grad student not a professor but I've been a TA or unofficial mentor for many electrical engineering and physics classes.
I'm sorry to see that the screening factor bothers so many humanities folks but I think it really is necessary.
For one thing, there's a minimum level of proficiency in solving physics problems that's required for me to be able to certify that you've "learned" introductory. electromagnetics for instance. Sadly there are lots of people who'll never reach that level of proficiency or will have to spend so much time and energy in doing so that they give up.
So what should we do differently? Make the class easier? Well then you're not really teaching the material. Break the class into multiple sub classes (so that intro E&M takes 2 semesters for instance)? This might be a workable solution but there's plenty to be taught in physics and limited time to do it.
I've worked a lot as an individual mentor helping regular undergrads and grad students as well as working with underprivileged secondary school kids. 1-on-1 instruction helps a great deal when teaching math/science. It's deeply rewarding to see kids who've been written off as unteachable master basic math and science. But at higher levels, I've found that even some very bright students simply never master the material. Perhaps this is just a failure on my part.
As for using intro science classes to educate laymen, my undergraduate university did this by having multiple levels of each intro course.
There was:
Physics 101/102 for laymen (mostly premed students),
Physics 103/104 for engineers and non-physicists in he physical sciences,
Physics 105/106 for physics majors (and masochistic engineers).
I don't know how exactly physics 101/102 were taught but I imagine the standards were a bit less severe
Neither your tenured faculty nor your graduate students are using intro level courses as an opportunity to teach the general public the importance of your field and its role in society and people's lives.
Teach the importance of my field rather than the field itself? I've never gotten the feeling that my physics students didn't know that physics was important.
Philosophers have to do this every day.
No comment.
Interestingly, I also spend a lot of time doing this for the hard sciences, because they don't do it themselves, and when they try to they do it badly.
How much of the material do you teach? Do you teach them how to solve problems using calculus for instance or do you not delve into math problems much? If you do, I'd be very interested in picking your brain about teaching techniques.
31: RHC, I think you misunderstand what I mean by "what they want, when they want to." I meant, when faculty teach 101/201 courses, they teach the class by selecting the concepts and their order while ignoring/defying the MCAT's suggestions. While I've stipulated the personal animus that former TAs like me felt towards pre-meds, I don't think that we viewed ourselves primarily as screeners, but as teachers. After all, we were only 2-3 years removed.
When I was a lab TA, I viewed myself as a cowboy on a cattle drive. It was my responsibility to get 12 students to the destination safely with as much learning (practice and theory) as possible. I didn't spend my time trying to wash out students. I know the profs at my institution weren't either.
As for the importance of the field, I think that chemists (my field) cannot help but emphasize the daily importance of chemistry. I don't know if we do a good job of it (there's way too much focus on explosions and poisons), but I think we do all right. I'm not surprised you do better / think you do better -- you probably do. But I have an easy one -- the physics and math folks have it much harder.
31:
Most STEM departments have, over the years, separated out GenEd intro courses from intro courses intended for prospective majors. That isn't the case in SS or Humanities departments. SOC 101 is both a course which satisfies the distribution requirements for a social science course AND the introduction to sociology for soc majors. A floor wax and a dessert topping. Math departments have special courses for SS and Humanities majors, so they don't have to take calculus. WillieStyle above (#37) is in a physics department where they have three separate two-semester "intro" sequences: one for the SS or Humanities major who has to take a two semester lab science course and has injudiciously chosen physics, one for non-physics STEM majors and one for physics majors. The course intended for people fulfilling the GenEd requirement may well try to "teach the general public the importance of your field and its role in society and people's lives". The others not so much.
I don't remember learning anything about the importance of the subject in my classes, whether it was chemistry or philosophy. The historical importance of different philosophers within philosophy, yes.
It seems unusual for a thread about teaching to be so dominated by hard-science people.
Not so sure about hard. I'm a little pudgy.
It's a mighty hard road from Lynchburg to Danville, a road with a three mile grade.
Never speak harsh words to your true lovin' husband?
This, of course, feeds in to another whine of mine about British education politics; for some reason the press always reports the number of "passes", i.e. exam grades A to G, or "passes A-C".
Cue homo craniorectalis trumpeting about falling standards, decline of the west, etc, etc. Of course, "pass" is an obsolete term - hell, anything below a C is very bad news - that had more relevance when not everyone *took* exams.
Equally, British degree classifications go as follows: fail, pass, third, 2:2, 2:1, first class honours. Frankly there is no way in which someone who gets a pass has "passed" in the sense of qualifying or successfully completing the course.
We should all quote grades as marks out of 100.