Home ยป Forum ยป Author Hangout

Forum: Author Hangout

Grading on the curve?

PotomacBob ๐Ÿšซ

For a story, (came up from am entry on the Story Score OP), who can explain "grading on the curve"?

Replies:   Uther Pendragon
Uther Pendragon ๐Ÿšซ

@PotomacBob

Literally, you're teaching a course, and youdecide before giving the test that the
10% best scores will get an A
20% next scores will get a B
40% in the middle will get a C
20% next scores will get a D
10% worst scores will get an F.

Even teachers usually modify that process; usually, they look for a sizable break in quality to give a break in grades.

As a metaphor, it's anything where you evaluate people in terms of what others in the same situation do.

Replies:   PotomacBob
PotomacBob ๐Ÿšซ

@Uther Pendragon

That sounds like it's judging students in a single class. I thought it involved grading based on multiple classes and maybe even multiple years - so you could find yourself competing with students from two or three years ago.
No?

Replies:   Dominions Son
Dominions Son ๐Ÿšซ

@PotomacBob

I thought it involved grading based on multiple classes and maybe even multiple years - so you could find yourself competing with students from two or three years ago.
No?

In my personal experience with it being used in high school and college, no it was usually just the single class.

Occasionally a teacher/professor might lump all their students in a given subject/level in a given semester together, but I've never encountered building a curve across multiple years.

Ernest Bywater ๐Ÿšซ

Uther has the details of how they do it, but it doesn't mean it's worth using in real life, despite many teachers and schools using it.

The biggest problem with grading on a curve is it has no way to show if anyone actually passed the test. I have seen end of year school results for some subjects where the top score was 45 out of 100 marks and the curve they used had the top 70% of marks passing, so the pass mark ended up being 20 out of 100 - which meant no one actually knew the material being tested in that exam.

Conversely, I've seen an exam where the lowest mark was 85 out of 100, and anyone with less than 93 out of 100 failed the exam.

Crumbly Writer ๐Ÿšซ

The idea, is that the teacher is grading the students based on how they compare to the other kids, thus is a very competitive environment, where whoever blinks first flunks and those leading the pack gloat, holding it over the heads of the others.

Mushroom ๐Ÿšซ
Updated:

Curves are one of the problems in education, and it has been a flaw for decades.

I now find this funny, but I was actually put in a "remedial reading" program when I started High School. I was put into it by a counselor who without even meeting me saw my grades and decided that I did not know how to read. So put me into a class designed to combat that, in which a student could not be pulled out to prevent incensed parents from their kids getting the help they really needed.

The joke? I was reading since I was 3. I freaked out my 6th grade teachers when as a book report, my presentation was on a book by Isaac Asimov. I showed up for my first class of "Remedial Reading" with a book I was currently reading, "Canterbury Tales", by Chaucer. In Old English.

Stuck in a class with kids that were in 10th and 11th grades, literally working through Curious George and Clifford, the Big Red Dog. I think the most advanced books in this entire class dedicated to reading was "Encyclopedia Brown".

Yea, easy A. But the downside I learned later is that nobody in that class can be placed in "good classes". I got classified in the computer system as a moron, so got in the moron classes. Classes where they normally used a curve because otherwise 75% of the students would fail.

Then along I come. Not low grades for anything other than I was bored. Take a test, get a 95, blow the hell out of the curve. Then the interesting aspect. Being in all the idiot classes, then getting called in by my counselor for the first time because she realizes I am acing "Computer Programming", the only non-idiot class I was in because it was so new it had not been put in the system for me to be kicked out of because they thought I could not read.

This was 40 years ago, and today it seems things are even worse. Kids who should be failed, given passing grades on a curve. Who then later enter the work force and fail because they had never learned how to become better.

Crumbly Writer ๐Ÿšซ

I suspect that this thread is getting distracted, largely because no one here know what PotomacBob was worried about when he first asked about grading on the curve. Was this in reference to a story about high-school classroom antics, or curiosity about a subject he read about but didn't fully understand. Knowing what was driving the question will allow us all to better address the question, rather than throwing out a growing list of alternative explanations.

Replies:   awnlee jawking
awnlee jawking ๐Ÿšซ

@Crumbly Writer

rather than throwing out a growing list of alternative explanations

In the thread I'm reading, nobody seems to think Uther Pendragon didn't nail it.

AJ

Replies:   Remus2
Remus2 ๐Ÿšซ

@awnlee jawking

In the thread I'm reading, nobody seems to think Uther Pendragon didn't nail it.

There are a lot of variations for how it's done, but in the main, he did nail it. Going on a tangent after that would not serve any good purpose.

Akarge ๐Ÿšซ

A couple of things not mentioned.
1 it's called 'grading on the curve' because it's based on bell curves.
2 It's statistics. Each grade A,B, C, etc is one or more standard deviations away from average. (C)

Dominions Son ๐Ÿšซ

@Akarge

That's the origin and the basic idea, but your #2 is not how it's generally implemented in practice.

richardshagrin ๐Ÿšซ

@Akarge

1 it's called 'grading on the curve' because it's based on bell curves.

There are many curves that may apply, the "normal" or "bell" curve can be less applicable with fewer data points to make it, or if the source of the data is not "normal". If all the students but one get 100% on a test, did the one with a 99% fail? There is a reason people who study statistics can earn PhDs. It can be complicated.

If you have to be one standard deviation from the mean to get a B and two standard deviations to get an A, there won't be very many A students, depending on where the mean is and how big the standard deviation is.

Replies:   bk69
bk69 ๐Ÿšซ

@richardshagrin

Class of forty students, two 'A's and two failures, most people get B, C, or D. Sounds about right.

I prefer the way my physics teacher graded tests. He'd mark mine, total the score, round to five, then check how many people failed. If too many, drop total by five and check again. Repeat as necessary. Yeah, gotta love having a >100% average.

Replies:   Ernest Bywater
Ernest Bywater ๐Ÿšซ

@bk69

Class of forty students, two 'A's and two failures, most people get B, C, or D. Sounds about right.

and that says absolutely nothing about how well they did or didn't know the material.

Years after I left high school I was doing a college course that had a statistics module and one of the things we examined was the various ways to mark exams. The prof doing the course had the raw exam results from three different high schools in the area where they all sat the exact same exams. We broke into 4 groups to analyse the results as: each class on a Bell Curve, the whole lot on a Bell Curve, each class by percentage of marks out of 100, the whole lot by percentage of marks out of 100. We got 3 different sets of results. In each of the groups where the marks were based on a percentage of the score the students all got the same grades with 5% getting an A grade with better than 90% of the marks, 75% of the students getting B grade as they got between 80 and 90% of the marks, and the rest got a C with the lowest mark being 75% of the marks possible. Yet on the Bell Curve we had many more C grades and a lot of D grades. That little exercise proved the stupidity of Bell Curve grading. The prof then showed how the previous semester's class did a similar exercise on another set of exams which showed the usual spread despite the highest mark being 45 out of 100, which was a fail of the exam by all involved.

Replies:   bk69
bk69 ๐Ÿšซ

@Ernest Bywater

and that says absolutely nothing about how well they did or didn't know the material.

True. But school systems regularly reduce the requirements anyhow. It's not like anyone really knows all that much if they've been 'educated' in public school. That's why the homeschool kids typically outperform public school kids in pretty much any academic competition.
Schools can't just go around failing kids, dontcha know, it might hurt their sensitive little feelings.

Replies:   Dominions Son
Dominions Son ๐Ÿšซ

@bk69

Schools can't just go around failing kids, dontcha know, it might hurt their sensitive little feelings.

Technically they are more worried about the parent's lawyers than the squirt's feelings.

Ernest Bywater ๐Ÿšซ

Almost forgot to mention, when grading on the curve make sure to have the road angled to the inside of the bend to make the water run off easier and the car run off harder.

Replies:   bk69
bk69 ๐Ÿšซ

@Ernest Bywater

That depends. NASCAR track curves, most city drivers would find their car sliding to the inside.

Grey Wolf ๐Ÿšซ

I have been in a class where the professor claimed (rightly) to be poor at writing tests. It was an honors Chemistry class. He grades with the A/B curve such that 50% of the students got A's. Overall, he was probably much more right than wrong; it would have been nice if his test-creation scores were correct, but it generally fit, and the Chemistry and ChemE majors didn't seem to suffer from not having lost the material.

On the other extreme, I had a professor in Psychology who curved to a bell curve regardless of the scores. She curved downward more than once; on one test a 95 was a B and a 91 was a C.

Back to Top

 

WARNING! ADULT CONTENT...

Storiesonline is for adult entertainment only. By accessing this site you declare that you are of legal age and that you agree with our Terms of Service and Privacy Policy.


Log In