What does that have to do with the question of why people don’t get 100% scores on the AP exam?
And AP exams are neither a model to follow for college nor a good comparison to normal distribution curves as given in a university.
@sylvan8798 “We get it. Humanities are the same as stem courses. No one course is any easier or more difficult than any other. It’s the new politically correct curriculum. Probably would help if you tell the student body that.”
That is very funny!
It does seem that no one is supposed to say that any major is more work than another or the pc police will jump up and challenge you. Different majors are not harder and easier, just different. You might think that you could find the more difficult majors, by finding the ones that students transfer out of, but you would be wrong. lol
Regarding the topic of “absolute” standards versus “relative” standards for a curve -
I think there’s a lot less difference between the two than most people think. In practice, many teachers use a blend of the two regardless of what their official policy is. For example, most teachers at my kids’ high school have a grading scale where A=90%, B = 80%, etc. In practice, if the kids do poorly on a test (i.e., too few A or B grades relative to “expectations”), teachers will often then give a fluff assignment for homework to pad the kids’ grades up. If they do well, then the next test gets made harder to bring the number of A’s and B’s back into line with the usual proportion. Relative curve.
In the colleges that I taught and TA’ed at, they usually used a relative curve. The professor would usually target a certain average grade, which ranged from B-'s to A’s. For humanities they usually targeted an A- or a B+; STEM classes would usually targeted a B-/B for lower division classes and a B+/A- for upper division classes (so humanities students almost always have higher GPAs than STEM students).
But in practice there’s an “absolute” component as well. If a class has less than say 15 students, a professor is usually assigning the final grade person by person based on their individual assessment of the student’s knowledge. Much more “absolute” in nature. For a class between 15-75 students, most professors would use the “official” proportions only as a starting point. Suppose the official curve was to give 30% A’s. They would look in detail at students on the A-/B+ borderline and decide whether these students did well enough (or poorly enough) to move the borderline up or down. The final proportion of A’s could be 25%, it could be 45% (usually professors give more A’s than they are “supposed” to). Same thing with the C+/B- borderline etc. They also use their own opinion as to whether or not a class is unusually talented to determine how to move the official curve around, as they would for very large classes with more than 75 students.
Even large non-STEM classes are graded on a curve. They just don’t bother telling the students that because they just slap a final letter grade on the papers and that’s all the students will see. When there are 200 papers to grade, they distribute them to the graders, telling that they should target (say) 40% or 45% A’s, etc. So students are “pitted against each other” for humanities classes too; they just don’t realize it as much. Humanities and (some) social science students don’t complain as much because i) the curve isn’t transparent ii) the grading is usually more generous. That’s all. It’s not because of “absolute” versus “relative” standards.
That’s why I don’t really think the “relative” vs “absolute” distinction is that meaningful, and I don’t get emotional about it.
Schools that flunk out 1/2 of their students from freshmen STEM classes are a whole 'nother kettle of fish. Got to decide … is it kinder to not admit them to the engineering school in the first place, or is it better to “give everyone a chance and see how they do”. Personally, I think it’s cruel and I understand why students and their parents get upset.
I think a lot of very good points are made in the this presentation. Good stuff.
There are also some problems with the usual grading scale of 90% = A.
When I went to elementary/high school the grading scale was (I think) 93% = A. I always thought it was stupid. A few silly arithmetic errors could cost you 3 or 4%. For me, getting a A meant wasting time on all the picune details that I had little patience for. I generally always got the same grade on math tests. If the test was super easy, I got a 95-97%. If the test was super hard, I got a 95-97%. I think using a 93% = A test just results in bad, somewhat unfair tests that don’t assess very well.
Personally, I was bored to tears and skipped school a lot … one main reason I went to school was so that I could play sports (you weren’t allowed to play in games if you were “sick” that day). I was one of those students who was both on the honor roll and in danger of flunking out. I was lucky that I had a high school teacher who intervened in my life. She gave me a bunch of her old college / graduate school textbooks, and worked out a deal where I wouldn’t have to go to 1/3 of my classes provided I read the books. Changed the course of my life.
@al2simon explained grading very well. It’s not as different as you think. Students often don’t fully understand what is going on. Although I can’t speak to humanities grading–that was pretty much a black box with a grade slapped on it unlike the STEM exams everyone is complaining about where you could clearly see that you were just plain wrong and exactly where you lost points.
@al2simon My friend who was my ds’s inspiring math coach when he was in high school tells a similar story about how 1 teacher changed her life by giving her college level math texts to take home and encouraging her. She ended up getting her PhD in math. 
If anyone is interested in a video of Richard Rusczyk’s talk, a link to it can be found here at the bottom of this page. (It won’t let me direct link bc my iPad doesn’t have flash player.)
@Pizzagirl = pc police might be the funniest thing I’ve read this week…
In my cynical moments I think professors in some other departments have the right idea. Make grading a black box and give lots of A’s so the students don’t complain. Wouldn’t have to spend lots of time coming up with original “tough but fair” problems for exams either. I think it’s irresponsible not to train and challenge students, but if you wanted to abuse the system it would allow you to spend more time on your research.
@mathyone of course it’s hard to solve problems. Everything is hard until you know how to do it. Even then it can be hard, but if you’ve been taught and you’re bright and you’ve put in the work it ought to be doable. And I do get that not everyone will put in the work out be able to do the work. If those are the ones failing, sure that’s fine. But it seems they could fail just the same on an absolute scale than because they are in a class of individuals who did relatively better.
A school’s strategy with respect to the above is largely determined by its admission selectivity, which is largely based on its popularity with prospective students, particularly highly qualified ones. A school can move the needle slightly on results (retention/graduation rates) based on academic support services and instructional methods, though some schools may try to game the reporting instead (see the “bunnies to be drowned” thread regarding Mount Saint Mary’s (Maryland)).
Of course, STEM is not just engineering; it includes natural sciences, particularly the biology and chemistry that includes the cutthroat-competitive pre-meds (and http://www.gradeinflation.com/tcr2010grading.pdf indicates that natural sciences tend to grade more harshly than engineering, which tends to have similar average grades as social sciences; all tend to have lower grades than humanities).
@OHMomof2 - The problems are “doable” but they aren’t always doable without a computer, with 4 or 5 other problems, in an hour, when you’ve just started working with the material 4 weeks earlier.
By the time the final rolls around, the first test material is now much easier because you’ve had 3 months more experience working with the ideas, seeing them in other contexts, making more mistakes … learning. I suppose we could wait to the end of the year when you expect mastery of the initial concepts before you start testing them, but then 1) you’d lose the learning opportunity that the 1st test provides and 2) nobody would know where they stood until finals.
@Much2Iearn It may not be PC, but I don’t think @sylvan8798 is necessarily incorrect. Here is a piece by Jonathan Wai on this very issue:
http://qz.com/334926/your-college-major-is-a-pretty-good-indication-of-how-smart-you-are/
@al2simon “(so humanities students almost always have higher GPAs than STEM students)”.
Why SHOULD humanities students get higher GPAs than STEM students? Can anybody tell me? Is there some philosophical or rational justification for this?
If Wai is correct (and I think he is), it is doubly obscene.
Btw, I am speaking as a social science grad.
@OHMomof2, Some students do better by virtue of natural ability and/or hard work. Why don’t all students get 100% in algebra 1? Diva algebra 1 teachers who delight in failing them? There is actually quite a large range of natural ability in STEM, and it requires a lot of hard work to make the most of one’s ability. As @dreadpirit pointed out, with more exposure, things become easier. And to a certain extent it’s true that we measure how fast students learn rather than whether they learn, which is unfortunate for the slower learners.
I agree with that, @Dreadpirit . First exam, sure. That’s an assessment to see how everyone is doing, much of the time, and the final usually accounts for more of the grade. ID the problem areas early so you can teach more effectively and students can study more effectively. Still see no need to create a curve that is designed to weed out, though.
@mathyone I am fine with the notion of grades that reflect the skills and work of the students. Just not as compared to each other.
Posting bits of quotes from the linked talk b/c I think it addresses many of the issues being discussed:
tyranny of 100% in high school
memorizing instead of mastering concepts
how problem-solving is taught
misplaced perception of calculus as the prime objective
Lots of good stuff in his talk. here is the pdf of the entire talk where i copied his quotes. http://mathprize.atfoundation.org/archive/2009/Rusczyk_Problem_Solving_Presentation_at_Math_Prize_for_Girls_2009.pdf
Being unwilling to create proper test problems that fairly evaluate your students is the mark of a bad teacher, plain and simple. Normal distribution curves are in my experience the sign of lazy teaching - the idea that instead of evaluating people on whether or not they achieved a given level of understanding of the material (a rather difficult thing to do), they simply come up with a ranking that is generally less than fair and pretend like it measures some degree of understanding of the material.
Also arbitrary and incorrect way of grading. There are proper ways of evaluating a paper against a rubric of what it’s supposed to have that, while subjective, are objective enough to generally give people the grade they deserve instead of arbitrary rankings.
In practice, with only one notable exception, the worst teachers all share the following qualities, which you have either explicitly or implicitly endorsed:
- A strong emphasis on their research work in such a way that it is to the detriment of their teaching. One particularly low-quality professor I had straight up said, “I don’t have time to care about your class because I have a multi-million dollar grant proposal that I need to work on.” It is not rare that terrible teachers care a lot about their research - if they had both terrible teaching ability and terrible research, they wouldn’t have lasted very long.
- An inability to write tests that are fair and of the proper difficulty. This takes a lot more work than just slapping a few questions together and calling it a test. Their tests often reek of lazy design and they don’t properly measure understanding of the material; there is a strong component of luck and randomness on lazy tests.
- An inability to properly gauge what the general level of understanding of their students should be on that test. Often but not always, this leads to a curve of substantial proportions that does not address the extent of student learning but rather just assigns letter grades and calls it a day. No “level of understanding” is measured in the process.
- A refusal to consider that this test may not have been properly designed. Can be justified with any number of the following excuses:
i. “We can’t lower our standards” (even though a normal distribution curve lacks any objective standard)
ii. “This is how it was done when I was a student and we didn’t complain.”
iii. “This is how it’s always been done.”
iv. “I did nothing wrong, my students are just stupid.”
v. (insert any other generic professorial excuse here)
Another tangential example is the fact that “normal distribution” rankings as done in universities have a measurably disastrous effect, in practice, when applied to companies. There’s no effort to measure how effective universities are at nurturing talent into high achievers, but given that academia has been in terminal decline for decades I suspect the effect may be the same as it is in the business world.
I think you envision that a bunch of students who have mastered the material are getting bad grades simply because they are being pitted against other students. I don’t think it’s true. Those students would also fare poorly under the kind of absolute grading scale you envision. And where would such a thing come from? Who gets to decide?
One of my kids is now taking an AP class which her sister took, but with a different teacher. Yes, they both used the exact same “absolute” and “objective” grading scale you all want to see, both teaching the same syllabus with the same texts, the same number of instructional hours, and very similar exam questions. Yet grades in this second teacher’s class are substantially lower, by about half a grade, because this teacher doesn’t grade-prop like the other teacher did.
Sorry if I re-cover some points…I read the first few pages of the thread and the last few.
A few things:
- STEM is harder for GPA because STEM is harder in real life. (I have a BS in a social science)
- Do you want doctors or other scientists that do OK? Or do you want the best and brightest?
- In real life you are compared to your peers in most jobs, not to an arbitrary standard…do you really want the whole class to fail? Or should the profs make it easier so everyone wins?
- Weeding out is important for the students on both sides. Do you want degree in STEM field and the subsequent job and then fail miserably at the job because you are simply not as good at it as the other people in the field? Better to learn early that it is not your best option and alter your path…Those that remain after weeding can focus on the nuance and intricate portions of their subjects rather than having to slow down for kids who are not doing well.
- Many of these bright student need to have the humility conferred by failing at something. S1 is an NMS in a STEM field…school was always easy until he ran into this type of system. Now he is making better choices about studying and applying himself because the challenge is there…getting As with more effort turns out to be much more satisfying…even if the ‘result’ was only a 50-60% success rate.
Thanks to whomever it was that pointed out that it is often outside of the class material and focuses on practical application. I have heard this complaint from S1 at times and this may help me help him understand why the class is that way rather than it simply being about ‘bad’ teaching.
“3) In real life you are compared to your peers in most jobs, not to an arbitrary standard…do you really want the whole class to fail? Or should the profs make it easier so everyone wins”
You’re not getting it. In the real world, a manager’s job is to get all - or at least the vast majority - of their subordinates to “master the material” and be strong performers. If I have people under me who consistently underperform, it’s my job to figure out how to help them learn, not just wash my hands of them and declare them metaphorical F students because I’ve determined a curve is more important.
“We get it. Humanities are the same as stem courses. No one course is any easier or more difficult than any other. It’s the new politically correct curriculum. Probably would help if you tell the student body that.”
Cool! So if I gave you the assignment of - from a blank sheet of paper, write a screenplay to be performed at the end of the year. Make it so that it touches people’s emotions. While you’re at it, write and arrange original music for it. Your “final” will be a performance that Broadway and other talent scouts will attend to look for who is the next new thing. That’s an actual “assignment” at my alma mater and they do it every year, brilliantly. Are you saying that you, as a STEM major, wouldn’t find that hard? You could do it with one hand tied behind your back, it’s that easy? Because I don’t buy it for a minute. I think many of you STEMmies find STEM easier for you than the humanities are, but you’d be flummoxed by the above assignment, or by comparing and contrasting Moliere to Corbeille, or by quite a lot of other things that don’t exist in a narrow conception of the world.