My D2 left campus last week to take the GRE. She didn’t have much time to study and was quite stressed about that. She ended up with very high scores, close to perfect in verbal and quant. She said, “Mom, I forget because my college is so hard that I am not stupid. It just makes me feel that way most of the time.” Now she loves her college, would not trade it… and fortunately she usually does okay in the academic self confidence department. But it made me a little sad to hear that. Who wants to feel unrelentingly like they are stupid for four years? Can’t we produce solid STEM graduates without doing that?
"What you have to understand about math/physics/engineering is what you mean by “the material”. It isn’t things like “Paris is the capital of France.” It isn’t just memorizing a bunch of formulas. It is being able to apply those formulas and concepts to various problems. "
Oh spare me. How is that different from the humanities? Did you think that history class was just merely “When did the Seven Years War occur” or that English class was “What was the name of Romeo’s sweetheart?”
The whole notion of curving is just ivory-tower bs as far as I am concerned. In the real world, a world that these professors don’t inhabit, it’s not as critical to “separate out” levels of performers according to a normal distribution curve. In the real world, a manager and a developer of people is supposed to try to make all their employees / charges “A” or “near A” performers, not arbitrarily fitting them along some curve. Of course, not everyone will be an A performer, but that reflects the bias of the real world which is “teach and inspire people how to do the work,” not “continually pit them against one another so the bottoms all drop out.”
I think a professor who gives a test where the average is 40% should come face to face with the fact that either he is giving a wildly too-hard test to satisfy his own ego, or he’s a cruddy teacher who is doing a poor job.
Re #100
But isn’t your daughter at a school known for extra rigor?
Yes and no. As managers, it’s critical to develop your team. However, at the end of the day, you still have to rate and rank each employee. This drives everything from merit pay and bonuses to who gets terminate in a force reduction. That’s the first thing I have to teach new managers. Everyone on your team can’t be a superstar. They can all be great, but some are going to be better…
Back to fantasy land (academia), if every student in a class is awarded an A, which ones have mastered the material better than the others? We don’t know. We don’t know who’s a good fit for engineering, or who’s a good Ph.D candidate. We do know, that everyone is not.
It’s an imperfect system for an imperfect world.
@ucbalumnus, yes. And they pile the workload on – they compress what are semester long classes at other schools into half a semester, load a lot of material into the classes, and require students to go farther in their non-major STEM subjects in the core classes than other colleges. But that still does not require a grading curve and teaching approach that makes kids feel dumb. There are still too many old school profs who haven’t changed their methods of teaching and testing since they got tenure 35 years ago.
Plus, this issue seems to exist in STEM at pretty much all colleges.
It is probably more of a convenience to instructors in that they do not have to finely calibrate the difficulty of new test problems to match an absolute grading scale. Of course, it only works in large classes where the distribution of student abilities tends to be similar from one semester to the next. In a small class, it is entirely possible that the one semester may have significantly stronger or weaker students than the previous semester, so relative curve grading has a greater chance of “inaccurate” grading.
While a test with 40% as a typical score may be a bit too difficult, why would it be a bad thing if the instructor writes a test that uses most or all of range of the possible scores for students who pass the course, rather than just the 70% to 100% range for students who pass the course?
For example, if the instructor writes a test with three equally weighted problems, one of which is an easy one that C students will solve, one of which is more difficult that B students will solve, and one of which is even more difficult that A students will solve, is it inherently bad if a class with many B students tends to have median scores in the 60% to 70% range (designated as a B letter grade)?
There’s nothing inherently wrong with that concept. I should have known you would have gone from conceptual to overly literal. The bigger point still stands. It makes real-world sense that the goal is that everyone masters the material to a high level. That’s the friggin’ GOAL of being a teacher / instructor / manager – to teach people how to do things. It’s only the academic ivory-tower nonsense that the goal is to be able to distribute performance along a bell curve. That shouldn’t drive how the process is done.
This is done in some companies and is know as stack ranking. It’s generally considered to be a disaster wherever it has been applied and comes with a lot of the kind of stupidity you should expect (political firings, lack of cooperation between employees, hiring “insurance incompetents” to fill the bottom of the totem pole, etc). It seems to come from academia because I see it more among companies that are the product of an academic environment. It’s also a product of the self-propagating idea of “hyper-competitiveness” that comes from those who did well in the academic system (not necessarily a group that overlaps with “high-achieving students”).
Generally the highly skilled people who lose this game, which is more political than meritocratic, either leave or become minimal performers. It’s a sign of a bad company if it pits its employees against each other.
Proper layoffs acknowledge that good, highly skilled people have to go because the economy doesn’t allow them to stay. And they rarely have all that much to do with merit but with politics and economics.
And if that A means that they met the standards for what they are expected to know, we shouldn’t care.
That’s exactly it, NeoDymium.
Let’s come out and say it - it’s nerd-ego to have to force a distribution to give only a certain number of A’s, B’s, C’s, etc.
We can but the truth is that there are those who don’t really want it to happen.
There are smart people who enjoy competition, and there are smart people who don’t. Both are probably equally capable but it’s the ones who are more competitive by nature tend to choose to play the academia/competitive admissions game. It appeals strongly to their innate sense of competition, what with all the prizes and awards and the individualistic nature of it all. These are the kinds of people who will generally go on to become professors, and especially so at top universities. Add on the fact that academic career advancement is dependent entirely on receiving your professors’ seal of approval (also known as a letter of recommendation) and you get a system that tends to be self-propagating and resistant to any form of change.
It’s usually grads from top schools who like to do it with little more justification than “this is how it was done when I went to school.” Lecturers from lower tier schools are less often like this.
I’m not sure a test is going to tell you who’d make a good engineer and who’d make a good PhD candidate, presumably both have mastered the material, the concepts, can think through the problems, etc. If the whole class earned an A, then I’d just think that means the class is full of intelligent students who worked hard and the teacher successfully taught them.
[quote]
@Pizzagirl It makes real-world sense that the goal is that everyone masters the material to a high level. That’s the friggin’ GOAL of being a teacher / instructor / manager – to teach people how to do things. It’s only the academic ivory-tower nonsense that the goal is to be able to distribute performance along a bell curve. That shouldn’t drive how the process is done.[/'quote]
I wish I could “like” this post more than once.
Then how is it that anyone can think in terms of “they only know 50% of the material” in ANY course? Personally, I don’t agree with you, as I am currently taking a French course where one either knows the meaning of “un pourboire” on the exam or one does not. One either knows when to use the subjunctive form and how to conjugate it or one does not. One can either write a paragraph describing how to get to the mall or one cannot.
It’s a tad different from physics thus far.
Wait until your French course starts grading you on how well you write in French. How clearly you articulate your ideas, how persuasive your arguments are, etc… @sylvan8798
I never grade on bell curves, but I do sometimes have to curve the grades upwards if the overall picture warrants. And just because the GOAL is that everyone master the material to a high level doesn’t mean that actually ever happens. Really - we can only do so much.
I definitely remember being graded on the quality of my speaking and writing, with success being measured on a spectrum, back when I learned Spanish in school. The entire idea of having a grade for the correct conjugation went away quickly after the beginners course.
Since most posters seem to be against “grading on a curve”, what is your opinion if the instructor grades on an absolute scale, but the thresholds were much lower then the ones used in high school, because a greater portion of the test is in difficult problems? I.e. suppose the test generally consists of three equally weighted problems:
one easy problem that a C student should be able to solve
one harder problem that a B student should be able to solve
one even harder problem that an A student should be able to solve
and the absolute grading scale was something like 25% = C, 55% = B, 85% = A (allowing for minor deductions on each problem). Would that be an undesirable testing and grading system?
Note: as used here, the definition of a C student is one who learns the material well enough to go on to the next course. B and A students are defined as those who can solve more difficult problems than the minimum expectation for passing the course. Obviously, the instructor would like it if all students learned the material well enough to solve the hardest problems, but that usually does not happen in larger classes.
I take issue with exams as tools to compare students to each other. If everyone in the class is great, everyone should get the great grade and move on. I think having a grade change based on who else happens to be in the class just makes no sense.
I suppose there’s a place for that sorting/weeding out, and in the scenario @ucbalumnus presents, the difference between students becomes clear without the bell curve. If everyone gets the “A” problem, then everyone gets an A.
Whatever. I think there’s a lot of chest-pounding and ego-driven nonsense in how you all talk about STEM and weed-out and so forth. If I were an instructor and my students mastered the material at A levels, then I’d give them all A’s if they deserved it. I think the professors who take glee out of sorting and having the bottom players drop out are lousy human beings.
@ucbalumnus I wonder the same thing.
One of the comments we received about our ds attending Bama vs. a top 20 school is that he wouldn’t be challenged and learn what he is capable of achieving. I am not 100% positive he is fully challenged. He says he is and that his grades reflect a lot of effort, so I can only accept his evaluation. But it is only being honest to say that I am thankful that his courses are not reduced to the low avg being the highest level of challenge in the class. I am glad at a state flagship he can have problems that are exceedingly difficult and may not be able to be solved by weaker students.
What the ratio of challenging questions should be, I have no idea. But for every struggling student, there are students who are advanced beyond the avg student in the class. It would be nice if they could all attend uber-challenging schools, but since that isn’t the way our society is designed, thankfully professors can raise the bar to meet that level of understanding. Otherwise, some skills that can be nurtured an developed are going to just go to waste.
But, how that differential works out in a college classroom is beyond my area of experience.
When I was in Ochem 1, one quarter dropped and one quarter failed.
Same thing happened in Chemical calculation for Chem. E class. I guess it’s a norm