[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math


View post   

File: 228 KB, 350x200, suicide.gif.webm [View same] [iqdb] [saucenao] [google]
11603095 No.11603095 [Reply] [Original]

>Calculus grade isn't curved

>> No.11603149

>>11603095
That would have been fine with me but my EE classes would have been a bloodbath without a grading curve.

>> No.11603157

This curving System is so utterly retarded.
American universities are a shit show except the very top.

>> No.11603164

>>11603095
What does curve mean again? I am European and I don't think we do this

>> No.11603169

>>11603164
https://en.wikipedia.org/wiki/Grading_on_a_curve
>For example, if there are five grades in a particular university course, A, B, C, D, and F, where A is reserved for the top 20% of students, B for the next 30%, C for the next 30%-40%, and D or F for the remaining 10%-20%, then scores in the percentile interval from 0% to 20% will receive a grade of D or F, scores from 21% to 50% will receive a grade of C, scores from 51% to 80% receive a grade of B, and scores from 81% to 100% will achieve a grade of A.

>> No.11603185

>>11603157
Took an introduction to algebra course at a community college. The instructor would adjust the grades to taking the square root of the earned grade and then multiply it by ten to get the recorded grade. You had to be pretty stupid to not get a passing grade in that class. A 36 average test score would equal a 60, which was a passing grade.

>> No.11603188

>>11603169
Why do all of this? We have a simple system where we grade to 10 and if you get 5 you pass (generally)

>> No.11603214
File: 6 KB, 163x241, file.png [View same] [iqdb] [saucenao] [google]
11603214

>>11603185
Goddamn. Talk about grade inflation. Was it a "hard" class (for somebody with no algebra experience, that is)?
>>11603188
My grade is hovering around 80%, but the class average on our last exam was ~55% so I think I'm near the top. The problem is it's going to tank my GPA like half a point.

>> No.11603217
File: 4 KB, 373x262, curvy.png [View same] [iqdb] [saucenao] [google]
11603217

>>11603095
>grade is curved

>> No.11603238

>>11603214
There were students in the class who struggled with the content but I don't know of anyone who failed. Community colleges don't have very high standards. Much of the class were education majors who were required to take one math course at the level of algebra or higher in order to get their associate degree.

>> No.11603241

>>11603214
I don't get why you guys use percentiles instead of absolute numbers. I get that it's possible that difficulty of the examination can very and feel subsections of a class feel that they have been tested harshly but this just seems like a pointless mess which serves no one but some hr or some similar sort of nonsense committee to maintain frame about the quality of education

>> No.11603247

>>11603241
*Vary and make

>> No.11603259

wow, did nobody get the joke? calculus.... curve..... har har

>> No.11603261

>>11603241
>I don't get why you guys use percentiles instead of absolute numbers.
It makes it easier when comparing exams with different numbers of questions.
15/19 = 79%, but also 19/24 = 79%

So then you can factor that in when weighing the grades.

>> No.11603267

>>11603241
what I like to do is make everything in my classes add up to 1000. Makes it easier for the students to understand what percentage/weight goes for each assignment. I teach undergrads though. so some amount of dumbassery is to be expected

>> No.11603273

>>11603261
But that only has meaning when your problems all give equal points and your exams are of equal or comparable difficulty. And even then, why not use a weighted system so your point total maxes out naturally to something like 10 20 40 50 etc? It's not like there won't be some relatively trivial direct applications of theory that can be written in 5 mins as well as some hard problems which may take half or even an entire hour

>> No.11603282

>>11603267
Didn't see your post before pressing. This is very logical and also helps you set a baseline as well as help your students make a plan on how to tackle the exam

>> No.11603351

>chaos theory grade depends mostly on initial answers

>> No.11603451

>>11603282
yeah I even do it for my 400-level and upper level undergrad classes, not just the 100's and 200's full of dipshits. lovable dipshits, but dipshits nonetheless.
even the "upper-level" kids shouldn't be forced to stare for hours at the syllabus trying to figure out how much a goddamn homework is worth so they can decide if it's okay to go out and get braindead with their friends or if they have to sit at home and finish a quiz. I'm a professor, not an asshole. molecular biology is hard enough for a beginner to grasp as it is

>> No.11603457

>>11603273
should've included this in the previous post, but I'd like to add. When I give out an easy low-risk homework or something, it's nice to point out that the 5 points means it's worth half of one percent. Exams are usually 150 points, 15% - just makes it all easy and clear and painless. life should be more easy, clear, and painless where possible. Especially science.

>> No.11603561

>>11603095
>grade curving
Do Americans really do this?

>> No.11603568

>>11603095
life isn't curved
get used to it big guy

>> No.11603917

>>11603164
it's the most retarded shit imaginable. thank god we don't have this burger shit
it means that if you get 90% of questions on a test right, but everyone else got higher than you then you fail. equally if everyone gets 40% and you got 45% you get a first class mark

I suppose it's a great motivation to not go to a prestigious university because then it's easier to get high grades since you have worse competition, but it's amazingly retarded that they would hand out bad grades to people who did well and good grades to people who did badly

>> No.11603966

>>11603561
kek I thought it was a meme at first. Always wondered about it when I realized it was real. Do professors get some extra pay or something if their class does very well?

>> No.11603972

>>11603095
Grading on a curve is fucking retarded.

>> No.11603975

>>11603966
>>11603917
>>11603164
The way my class did it was
>almost everybody bombed the first exam
>guy with the highest score was a 92
>you were allowed to "redo" your exam as a take-home thing, if you explained where you went wrong
>you could make up to 8 points from doing this
it wasn't much but brought my grade from a 79 to an 87

>> No.11604008

Is grading on a curve based on histogram equalization? i.e. your score is converted into a percentile, which gets replaced by the equivalent score for a pre-selected normal distribution. Or is the curve based on switching to a normal distribution by matching mean and variance? Because if it was the former then you would lose a lot of information about student performance. For example, a bimodal distribution in scores due to favoiritism or background or something.

In Canada when profs said they'd curve something they usually just meant multiplying the grade by some constant to adjust the mean.

>> No.11604202

>>11604008
>>11603975
my university had an interesting way of adjusting grades when students did much worse than expected on an exam. Not sure if this is common or not, but they would calculate the amount everyone's grade deviated from that person's average grade in previous exams and then multiplied everyone's grade by a constant factor set to make the total deviation equal 0. I.e. if there are [math]N[/math] people in the class and person n has unadjusted grade [math]x_n[/math], adjusted grade [math]x_n'[/math] and person n's average grade is [math]\bar{x_n}[/math], then
[eqn]
x_n \to x_n' = \min(k x_n, 100\%), \hspace{1cm} \text{$k$ such that } \sum_n^N \bar{x_n} - k x_n = 0
[/eqn]

>> No.11604266
File: 40 KB, 640x628, eyeroll.jpg [View same] [iqdb] [saucenao] [google]
11604266

>>11603095
>/sci/ is full of middle-schoolers who haven't even passed ODEs yet

>> No.11604267

>>11604008
>they usually just meant multiplying the grade by some constant to adjust the mean.
I would be fine with this t b h. Assuming they wanted to bring the mean to a 75%, the multiplier for this course would be ~1.3 and give me an A

>> No.11604297

>>11603169
What if every student in the class scores above 90%? Does that mean some students scoring say 91% are gonna get a D? that's fucked mate. wow americans are dumb.
>punished for having smart classmates
>rewarded for having dumb classmates
MURRICA FUK YEA

>> No.11604356

>>11603157
It is necessary because a lot of the professors are autists and either think an exam far too long and hard to be completed in the allotted time is fair, or they purposely design their exams to be like that so they can fail the whole class and transform the grades to whatever they want.

>> No.11604382

>>11604297
No one does anything like that anymore. Universities are in the business of collecting tuition and fees. They can't collect money from students who have been kicked out so there's a lot of pressure to never fail anyone. You have to actively try to fail a class now.

>> No.11604456

>>11603157
The very top is worse due to inflation

>> No.11604462

>>11604266
ODE's are easier than integrals to me. idk why

>> No.11604490

>>11604462
probably because you've only ever encountered ones from the tiny subset of ODEs that are exactly solvable. you won't get very far trying to solve something like
[math]y + (y')^2 + 2y''=0 [/math]

>> No.11604510

If a few students are confused, they're bad students. If the whole class is so confused that you need to curve grades, you're a bad teacher.

>> No.11604527
File: 457 KB, 598x478, Rich.png [View same] [iqdb] [saucenao] [google]
11604527

>>11603095
Calculus is not hard.

>> No.11604761

>>11603273
If you're in the business of coming up with exam questions, you'll find that some problems you come up with are unexpectedly difficult to people other than yourself. Unless all your exams follow the exact same formula (i.e. your exams are easy to pattern match for anyone who has an exam bank) this means that year to year, your exams are going to have varying difficulty. Sometimes wildly varying.

If you take the axiom that your student body is statistically diverse and aren't changing year to year, then grading on percentiles would serve to normalize scores across years.

It's similar to measurements of IQ, I guess.

>> No.11604781

>>11603157
Yeah....okay
.>>11603169
It is never practiced so absolutely - whatsoever. That's just a theoretical summary.
>>11604297
Then they all get A's/A+'s. There is no university where curves are applied consistently or indiscriminatorily. You just have a small brain and a poor ability for abstract thought or consideration. Hence why Americans can, and always have been able to, destroy and blow you up.

>> No.11604790

>>11604490
This is very easily solved with a fourier transform.

>> No.11606076

>>11604790
do it then

>> No.11607171

>>11603917
this is the most retarded interpretation i'ever seen.
> if you get 90% of questions on a test right, but everyone else got higher than you then you fail
do you really believe this? it doesn't matter if everyone did better or worse. all that means is if the highest was, say, 96, you will get 90+4=94.
if someone got 100% you won't be curved. but there is no way you are going to fail simply because someone did better. Also that's just one particular kind of a curve (a flat curve). There are others, for example ax/b + c where everyone's result x is curved. As long as f(a)>f(b) if a > b to prevent leapfrogging.
And most importantly x <= f(x) to guarantee that your curved grade cannot be worse than the raw result.

>if everyone gets 40% and you got 45% you get a first class mark
you have no idea what you are talking about

>> No.11607312
File: 54 KB, 872x960, FE6F45C2-A9D7-4D3B-9443-E7609BEF4F80.jpg [View same] [iqdb] [saucenao] [google]
11607312

This was the grading scheme for a stochastic processes course I took. The material was a little challenging but it was impossible to fuck up the work

>> No.11607321
File: 55 KB, 855x586, inflate.png [View same] [iqdb] [saucenao] [google]
11607321

Interesting article on grade inflation.
https://www.cbsnews.com/news/grade-inflation-colleges-with-the-easiest-and-hardest-grades/
It appears most of the Ivy League, except for Princeton, are guilty of pumping up grades.
>At private schools the average GPA is 3.30 and at some of these schools the average is 3.5 or even 3.6. At Brown University, two thirds of the grades are A's.

>> No.11607327

>>11603568
>taxes, social security, minimum wage, etc. aren't life curves

>> No.11607377

>>11603169
I've never actually had nor heard of a class that grades on a curve such that it's guaranteed x% of the class fails every semester. What's more common is that there's a set amount of dropped points determined at the start of the semester or a flat curve determined at the end.

>> No.11607601

>>11607321
is this because of inflation or just better students/environment to learn? i mean most people getting into these top universities have already shown they can study hard and do well in school, and these top school usually have smaller classes and more time with TAs or professors which is all conducive to a better grade. maybe it talks about that in the link and they factor for that though

>> No.11607625

>>11607601
Two decades ago your theory might have been worth considering but schools, especially Ivy League, admit students on diversity points more than anything else. Selectivity is gone, everyone who wants to go to college gets to do so, regardless of their academic abilities. The average student of today is quite inferior to the average student of two or three decades ago.
Grade inflation is from several things. 1) Schools don't make money off of students who have flunked out 2) Diversity requirements push for passing anyone on the oppression totem pole and 3) Snowflakes who threaten to sue if they don't get the grade they want. Most professors find it easier to just hand out high marks rather than deal with all the crap that comes from honest grading. The school administration wants the higher grades so they're not going to be a check on grade inflation either.

>> No.11607690

>>11604781
>Then they all get A's/A+'s. There is no university where curves are applied consistently or indiscriminatorily. You just have a small brain and a poor ability for abstract thought or consideration. Hence why Americans can, and always have been able to, destroy and blow you up.
retard

>> No.11607719

>>11603188
Some professors are shit at teaching and evaluating the difficulty of their exams. In my upper div math classes in uni, the uncurved medians of the midterms and finals were almost always below a 70. The lowest I've seen was around a 55. I know that many physics and EE classes also have the same problem.

>> No.11607729

>>11607719
Where do you think the blame lies? Bad teaching? Bad tests?

I've only been hitting low 80s in my exams because I just run out of time and end up rushing. Yet somehow I'm near the top of the class in grades....

>> No.11607730

>>11607690
European arguments

>> No.11607749

>>11607729
I think, in my case, it's both the professors and the students. I went to a fairly prestigious research university for undergrad, so the professors there weren't focused on teaching us, leading to bad teaching and bad tests.

On the other hand, a good chunk of the students in the math department were interested in pursuing CS instead of math, but couldn't switch majors due to it being impacted. From the few math/CS people I knew, they'd often put more time and effort into their CS classes instead of math.

>> No.11607762

the problem with undergraduate calculus courses is that
A) the professors are usually {adjuncts (foreigner who can't speak English) or old fogies on their way out} = poor quality
B) many students are on unequal levels due to high school courses. some kids have already taken up to calc III in hs and are smurfing in classes that are just review for them, throwing off the professor's assessment of the students' progress.

i hated those courses

>> No.11607794

>>11607762
My uni has their own site for course evaluations and I always saw the Calc 1 or Precalc classes for non-STEM people giving out the lowest grades each semester, like a C-/D+ average. I always thought it was just retarded non-STEM students only taking it for a GE, but the shit adjunct profs were also probably a factor. I've even recognized some of my past TAs teaching those classes, and they weren't always the best. I never took Calc 1 or 2 in uni and most people there recommend not taking them as refreshers.

>> No.11607799

>>11607762
that was me for calc 1-3, learned a lot that was covered in high school already and got easy As. diff eq was similar enough to get an A as well but linear algebra was a little different and being lazy fucked me. i probably should have taken the AP tests in high school but i had a tuition scholarship for uni so it was just free gpa boost

>> No.11608444

>>11603259
Don’t know if it was intentional but that was my first impression too lol

>> No.11608735

>>11604790
Yiff in hell, fourriers

>> No.11609369

>>11603185

>Introduction to algebra course in college
I wouldn't even curve that. If you can't get above 70% (C- under standard grading scale), you pretty much deserve to fail.

>60, which was a passing grade
But it's a D-. It's a failing grade.