whats the point of this GPA stuff?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

SaturnX

Diamond Member
Jul 16, 2000
3,415
0
76
Yeah, I very much prefer percentages, it's just easier to understand... but then again so is Celcius, haha, so I guess it just has to deal with how we're brought up in Canada.

--Mark
 

Playmaker

Golden Member
Sep 17, 2000
1,584
0
0
Originally posted by: Colt45
when I was in highschool here, we never used GPAs. I'm not sure if all of canada is like this? or maybe just the west..


Anyways all they gave us was a straight percentage. no weighting or any of that.

What is the advantage of 3.5 over 86.2% or whatever? I don't see the point.



When I was in technical school, it was the same thing, no GPA. (although they would curve it if the average was brutal)


So.. someone want explain the purpose of it to me?

Welcome to the inefficiencies and stupidities of academia. What's really ridiculous are the schools that don't even follow the standard 4.0 system, but develop some convoluted system of 5.0 for an AP course A in high school or 4.33 for an A+ in college, throwing everything off. Going to a college where 4.0 is the highest grade, but there are + and - s for every other grade is disparaging when applying to law school, as at a school on the 4.33 system, half of my As would be A+s giving me a substantial GPA boost. The same thing happened with AP courses in high school...And this system was supposed to develop a STANDARD for comparison?

Why not move to a standard 90A, 80B, 70C, 60D system for every class and cumulative grades? It would just make too much sense, and that's not an area in which institutions of learning excel.
 

BlueFlamme

Senior member
Nov 3, 2005
565
0
0
And not all schools in the US are on the ten point scale. My district had the following screwball setup:

A = 100-94
B = 93-85
C = 84-75
D = 74-65
F = 64-0

No plusses or minuses for GPA, so you could have a 93.4 and share a 3.0 with the person who just squeaked in at 84.5.

And to all the schools who weight for AP courses, it is out of hand. I've heard of people with 5.x GPAs.

Not to mention in college there is no A+, which they rationalize by saying there is no F- :roll:
 

dullard

Elite Member
May 21, 2001
26,066
4,712
126
Originally posted by: Lonyo
That's not the goddamn question.

WHY BOTHER MAKING IT INTO ANOTHER NUMBER?!?!
The point is there's no need to take say 75% and make it into 3.0 when you can just use the 75% instead.
Sure, colleges may use GPA, but using a raw % is just as easy, that's the point.
You just don't understand. In the US, it starts at 3, why bother to make it into another number like 75%?

 

mugs

Lifer
Apr 29, 2003
48,920
46
91
Really, they don't make much sense. They only serve to decrease precision and proportionality of grades. If grades were given solely as letter grades, then the 4 point scale provides a convenient way to translate those grades to a number. But nearly all teachers use percentage-based grades and translate that to a letter if necessary.
 

ManSnake

Diamond Member
Oct 26, 2000
4,749
1
0
Originally posted by: dullard
Is it that hard to multiply by 0.25? Or if multiplying is too difficult, is it that hard to divide by 4? Do we really need a rant for this?

3.0 * 0.25 = 0.75 = 75% = B average

No that's not correct. GPA is the cumulative average of one's grades for all classes. Grade 'B' has a value of 3, but someone might have received all 85% in all his classes which translates to letter grades of 'B', his percent average would be 85%, but his GPA would still be 3. Thus your calculation does not work.
 

mugs

Lifer
Apr 29, 2003
48,920
46
91
Originally posted by: dullard
Originally posted by: Lonyo
That's not the goddamn question.

WHY BOTHER MAKING IT INTO ANOTHER NUMBER?!?!
The point is there's no need to take say 75% and make it into 3.0 when you can just use the 75% instead.
Sure, colleges may use GPA, but using a raw % is just as easy, that's the point.
You just don't understand. In the US, it starts at 3, why bother to make it into another number like 75%?

In the US it starts at 3? :confused: Every professor I had in college kept grades as percentages. Then they converted that to a letter, and that was converted to a number. But whether I got a 90, 91, 92, or 93, that number was a 3.667. That's silly.
 

mercanucaribe

Banned
Oct 20, 2004
9,763
1
0
Originally posted by: OdiN
Originally posted by: Colt45
when I was in highschool here, we never used GPAs. I'm not sure if all of canada is like this? or maybe just the west..


Anyways all they gave us was a straight percentage. no weighting or any of that.

What is the advantage of 3.5 over 86.2% or whatever? I don't see the point.



When I was in technical school, it was the same thing, no GPA. (although they would curve it if the average was brutal)


So.. someone want explain the purpose of it to me?


3.5 would be 87.5%.

Just divide 3.5 by 4 to get the percentage. It's really the same thing...just expressed in a different way. I don't know what you are talking about with the weighting thing and all.

2.5 GPA = 62.5% = D

A C+ = 79% = 3.16 GPA

I'm confused as to what you are confused about.

No... that's not how GPA works.
Someone makes the following grades: 91 92 80 80 and has a GPA of 3.50.
Someone makes the following grades: 98 97 89 79 and has a gpa of 3.25 even though the average is much higher.
 

dullard

Elite Member
May 21, 2001
26,066
4,712
126
Originally posted by: mugs
In the US it starts at 3? :confused: Every professor I had in college kept grades as percentages. Then they converted that to a letter, and that was converted to a number. But whether I got a 90, 91, 92, or 93, that number was a 3.667.
Originally posted by: ManSnake
No that's not correct. GPA is the cumulative average of one's grades for all classes. Grade 'B' has a value of 3, but someone might have received all 85% in all his classes which translates to letter grades of 'B', his percent average would be 85%, but his GPA would still be 3. Thus your calculation does not work.
You both are talking about something completely different.

If you have these grades: A, B, B, D. What is your GPA? In the US, typically the answer is (4 + 3 + 3 + 1) / 4 = 2.75. Why bother to divide this by 4 to get 68.8%?

What you are talking about is not calculation of GPA, but calculation of the original grades (the A, B, B, and D). That is a completely different idea. There are several good reasons to convert to letters instead of using percentages in the grades. We can go into this side topic if you wish. But we are talking about converting the A, B, C, D, and F into a number.
 

mugs

Lifer
Apr 29, 2003
48,920
46
91
Originally posted by: dullard
Originally posted by: mugs
In the US it starts at 3? :confused: Every professor I had in college kept grades as percentages. Then they converted that to a letter, and that was converted to a number. But whether I got a 90, 91, 92, or 93, that number was a 3.667.
You are talking about something completely different.

If you have these grades: A, B, B, D. What is your GPA? In the US, typically the answer is (4 + 3 + 3 + 1) / 4 = 2.75. Why bother to divide this by 4 to get 68.8%?

What you are talking about is not calculation of GPA, but calculation of the original grades (the A, B, B, and D). That is a completely different idea. There are several good reasons to convert to letters instead of using percentages in the grades. We can go into this side topic if you wish. But we are talking about converting the A, B, C, D, and F into a number.

I'm not talking about something completely different, I'm talking about exactly what the OP is asking, which is why we use GPA instead of the actual percentage grade [edit: that is, the percentage grade that the GPA is ultimately based on, by way of the letter grade]. I think you and Odin were the only people who brought up the idea of dividing by 4 to get to a percentage, and Odin retracted that statement.

So it's not a side discussion, this is the original discussion. What is the merit of starting with a number grade, say an 89, then converting that to a letter, B+, then converting that back to an entirely different number, 3.3333? Why should someone who got an 87 also get a 3.3333, when they got a lower grade in the class?
 

dullard

Elite Member
May 21, 2001
26,066
4,712
126
Originally posted by: mugs
I'm not talking about something completely different, I'm talking about exactly what the OP is asking, which is why we use GPA instead of the actual percentage grade. I think you and Odin were the only people who brought up the idea of dividing by 4 to get to a percentage, and Odin retracted that statement.

So it's not a side discussion, this is the original discussion. What is the merit of starting with a number grade, say an 89, then converting that to a letter, B+, then converting that back to an entirely different number, 3.3333? Why should someone who got an 87 also get a 3.3333, when they got a lower grade in the class?
Original question:
Originally posted by: Colt45
What is the advantage of 3.5 over 86.2% or whatever? I don't see the point.
That clearly asks what I was discussing (it says nothing about converting the homework + quizzes + tests into letter grades). But I will go into this side topic for you.

Difference in difficulty
Imagine a university with two professors teaching the same course. Both professors use the same book, cover the same material, everything is the same. Students randomly are assigned to each professor so the classes are equal in ability. Along comes the final.
[*]Professor X chooses to make the final a piece of cake. All of his/her students score 80% or above on the final. The students who did poorer in the class scored near 80%, the students who did average in the class scored near 90%, and the students who did well in the class scored near 100%.
[*]Professor Y chooses to make the final a very difficult final, one that no student could possibly finish in the given time. There are reasons for this (to really separate the A students from the A+ students, since the professor is mean, since the professor is from a country that customarilly does that, since the professor accidently gave too many problems, etc.) The students got between a 20% and an 80% on that final. The students who did poorer in the class scored near 20%, the students who did average scored near 50%, and the students who did well in the class scored near 80%.
[*]Think about the average student in both classes. Both of these average students know the same material and would score the same if they took the same final. Should their class grades be the same? I say yes. Your method says NO. The student with the easy final would have a much higher percentage than the student with the difficult final. Is that fair? No. Converting to a letter system virtually eliminates this unfairness. A 90% on the easy final is a C. A 50% on the hard final is a C. Both average students know the same material, and get the same letter grade.

Difference in grading systems
Do the same thing with two different university systems. In some universities a 50% is passing in many other universitities a 50% is failing. Why should a student who did average (50%) in one university with a C being 40% to 60% be treated worse than a student who did average (75%) in a university where a C is 70% to 80%?

Difference in grading accuracy
Could you honestly tell me the difference between a 50 page essay that gets a 81% and a 50 page essay that gets a 82%? Probably not. Professors can't realistically grade to that level of accuracy with complete fairness. Converting to letters allows for some wiggle room. Both papers deserve the same grade (probably a B) because it is impossible to fairly assign percentages in that case.

Have you ever seen a case where two students did the same thing and got slightly different grades? I've seen that happen in almost every course I've been in. A professor/teaching assistant might subtract 6% for an error on the first assignment, yet after grading 200 assignments, the professor/teaching assistant might subtract 7% for that same error for another student. Sometimes it is very difficult to remember exactly how much was taken off for every error on every assigment. This is especially true when you grade on multiple days with multiple teaching assistants. So does the student who got 7% taken off for the same error deserve to be marked lower forever? No. Coverting to a letter grade eliminates these minor differences. Both students get the same grade and the same GPA for doing the same work. A rubric would help, but you cannot predict all possible student errors before, so the rubric is only partially helpful here.
 

mugs

Lifer
Apr 29, 2003
48,920
46
91
Originally posted by: dullard
Originally posted by: mugs
I'm not talking about something completely different, I'm talking about exactly what the OP is asking, which is why we use GPA instead of the actual percentage grade. I think you and Odin were the only people who brought up the idea of dividing by 4 to get to a percentage, and Odin retracted that statement.

So it's not a side discussion, this is the original discussion. What is the merit of starting with a number grade, say an 89, then converting that to a letter, B+, then converting that back to an entirely different number, 3.3333? Why should someone who got an 87 also get a 3.3333, when they got a lower grade in the class?
Original question:
Originally posted by: Colt45
What is the advantage of 3.5 over 86.2% or whatever? I don't see the point.
That clearly asks what I was discussing (it says nothing about converting the homework + quizzes + tests into letter grades).

Maybe Colt45 could clarify what he's asking, because I think it's pretty clear from that post (3.5/4 != 86.2) and this one:
Originally posted by: Colt45
86.2 was just a random number ;)

Im confused because it is based off the percent. so wtf is it's purpose? just a useless buffer as far as i can see

that he's not talking about conterting the 4 point grade to a 100 point scale, he's asking why we don't stick with the original 100 point grade to begin with.

But I will go into this side topic for you.

Difference in difficulty
Imagine a university with two professors teaching the same course. Both professors use the same book, cover the same material, everything is the same. Students randomly are assigned to each professor so the classes are equal in ability. Along comes the final.
[*]Professor X chooses to make the final a piece of cake. All of his/her students score 80% or above on the final. The students who did poorer in the class scored near 80%, the students who did average in the class scored near 90%, and the students who did well in the class scored near 100%.
[*]Professor Y chooses to make the final a very difficult final, one that no student could possibly finish in the given time. There are reasons for this (to really separate the A students from the A+ students, since the professor is mean, since the professor is from a country that customarilly does that, since the professor accidently gave too many problems, etc.) The students got between a 20% and an 80% on that final. The students who did poorer in the class scored near 20%, the students who did average scored near 50%, and the students who did well in the class scored near 80%.
[*]Think about the average student in both classes. Both of these average students know the same material and would score the same if they took the same final. Should their class grades be the same? I say yes. Your method says NO. The student with the easy final would have a much higher percentage than the student with the difficult final. Is that fair? No. Converting to a letter system virtually eliminates this unfairness. A 90% on the easy final is a C. A 50% on the hard final is a C. Both average students know the same material, and get the same letter grade.

My professors handled that situation by adjusting the grades on each test to a different number grade. A 50% on a difficult final might be adjusted up to an 80%. Using letters doesn't really offer any advantage there.

A C is only comparable to a C if every professor uses a bell curve. That's not the case. Same with a number grade of course, but if there's no advantage to converting to a letter grade you might as well stick with a number...

Difference in grading systems
Do the same thing with two different university systems. In some universities a 50% is passing in many other universitities a 50% is failing. Why should a student who did average (50%) in one university with a C being 40% to 60% be treated worse than a student who did average (75%) in a university where a C is 70% to 80%?

Can you really compare a C at one university to a C at another and say they're the same level of achievement? No. I've never heard of a 40% being a C. :confused: Except of course in the case where a curve is being used, which was your first point.

Difference in grading accuracy
Could you honestly tell me the difference between a 50 page essay that gets a 81% and a 50 page essay that gets a 82%? Probably not. Professors can't realistically grade to that level of accuracy with complete fairness. Converting to letters allows for some wiggle room. Both papers deserve the same grade (probably a B) because it is impossible to fairly assign percentages in that case.

You also couldn't tell the difference between a final grade of 79 and 80, but in terms of GPA there's a sizeable gap.

Have you ever seen a case where two students did the same thing and got slightly different grades? I've seen that happen in almost every course I've been in. A professor/teaching assistant might subtract 6% for an error on the first assignment, yet after grading 200 assignments, the professor/teaching assistant might subtract 7% for that same error for another student. Sometimes it is very difficult to remember exactly how much was taken off for every error on every assigment. This is especially true when you grade on multiple days with multiple teaching assistants. So does the student who got 7% taken off for the same error deserve to be marked lower forever? No. Coverting to a letter grade eliminates these minor differences. Both students get the same grade and the same GPA for doing the same work. A rubric would help, but you cannot predict all possible student errors before, so the rubric is only partially helpful here.

Same as above.

The fact is, a GPA from one school is not comparable to a GPA from another school, that's why they use things like class rank and standarized tests. Class rank is still not really comparable between schools, but it is as comparable as a GPA would be in an ideal world.
 

dullard

Elite Member
May 21, 2001
26,066
4,712
126
Originally posted by: mugs
My professors handled that situation by adjusting the grades on each test to a different number grade. A 50% on a difficult final might be adjusted up to an 80%. Using letters doesn't really offer any advantage there.
Letters do not offer any disadvantage. Letters are quick and easy to understand. A grade of B is easilly recognized and understood. A number of 80% that was adjusted up from 50% since test number 3 was 25% more difficult than test number 1 is just plain confusing. So was that 80/50/3/25/1 a good score or a bad one?
A C is only comparable to a C if every professor uses a bell curve. That's not the case. Same with a number grade of course, but if there's no advantage to converting to a letter grade you might as well stick with a number...
There are plenty of ways to give a C meaning without using a bell curve. When I tought my course (and graded for courses), I had a set of things I wanted each student to know. If they knew it, they got at least a C. If they didn't know them, they got a D or an F. In some classes, everyone knew the critical data, and everyone got a C or better. In other classes, many didn't know it and many failed. A bell curve is not needed at all.

You didn't address my issue though, you just changed the subject here. Why should a student in the hard class get a low number and the student in the easy class get the high number? Yes we can artificially adjust numbers, but then why not just artificially assign letters that are easy to interpret?

Can you really compare a C at one university to a C at another and say they're the same level of achievement? No. I've never heard of a 40% being a C. :confused: Except of course in the case where a curve is being used, which was your first point.
Why not? A student who fails at school X is comparable to a student who fails at school Y. (Replace fail with any other grade if you wish). They covered the same material and showed the same level of proficiency. If you wish, you can give more credentials to some schools, but a student who gets all F's at MIT is probably not the same as a student who gets all A's at South Dakota School of Mines. You can compare grades within limitations. I would give a C average at both schools the same treatment. But the graduation from MIT would put that person above the graduation at SDSM.

As for 40% being a C, you need to look at foreign schools. It happens all the time. In many countries, getting above a 50% is a real accomplishment. Should these people be punished for being from a country where 50% is great? No. Instead, artifically convert the numbers into a relatively consistant system. Is it flawless? No. But at least you can say a person with a B average is a decent student - no matter what country they are from. You can't say the same about a student with a 60% average. Is that good or bad? Without knowing the country and detailed knowledge of the country's gradings system the 60% is meaningless.

You also couldn't tell the difference between a final grade of 79 and 80, but in terms of GPA there's a sizeable gap.
True, it doesn't fix everything. It fixes 90% of the problems, but on the gaps there are issues. But anytime you are on the border of a class, you should talk to the professor and get things straightened out. But if you really look at the whole scheme of things, a C+ vs a B- in one class has no significant impact on your GPA (we are talking about ~0.01 point difference in GPA). And no employer cares about 0.01 points.

The fact is, a GPA from one school is not comparable to a GPA from another school, that's why they use things like class rank and standarized tests. Class rank is still not really comparable between schools, but it is as comparable as a GPA would be in an ideal world.
GPA is one of many statistics needed to compare people. Think of these two people, X and Y. Both are identical in every way. Except on one minor assignment on one class, person X got 6% taken off for a mistake. Person Y got 7% taken off for the same mistake. With GPAs, they both have the same class rank. With raw percentages, they have different class ranks with person X ahead of Y. GPAs do help in many of those situations.
 

mugs

Lifer
Apr 29, 2003
48,920
46
91
Originally posted by: dullard
Letters do not offer any disadvantage. Letters are quick and easy to understand. A grade of B is easilly recognized and understood. A number of 80% that was adjusted up from 50% since test number 3 was 25% more difficult than test number 1 is just plain confusing. So was that 80/50/3/25/1 a good score or a bad one?

Letters do have a disadvantage, there are fewer possible grades that can be achieved. As I pointed out, a difference of one point in a person's grade could affect their 4-point equivalent by .333 or it could not affect it at all. A person who just missed getting an A gets the same letter grade as someone who just barely made an A-.


There are plenty of ways to give a C meaning without using a bell curve. When I tought my course (and graded for courses), I had a set of things I wanted each student to know. If they knew it, they got at least a C. If they didn't know them, they got a D or an F. In some classes, everyone knew the critical data, and everyone got a C or better. In other classes, many didn't know it and many failed. A bell curve is not needed at all.

And is that inherently comparable to what another professor would score a C or a D or an F? Doubtful. So there's no benefit to choosing that vs. a number. A number on the other hand would let you distinguish between someone who knew just barely enough to get a C and someone who easily had a C but did not quite have the next higher grade.

You didn't address my issue though, you just changed the subject here. Why should a student in the hard class get a low number and the student in the easy class get the high number? Yes we can artificially adjust numbers, but then why not just artificially assign letters that are easy to interpret?

How are letters any easier to interpret than a number? :confused: People understand percentages.

Why not? A student who fails at school X is comparable to a student who fails at school Y. (Replace fail with any other grade if you wish). They covered the same material and showed the same level of proficiency. If you wish, you can give more credentials to some schools, but a student who gets all F's at MIT is probably not the same as a student who gets all A's at South Dakota School of Mines. You can compare grades within limitations. I would give a C average at both schools the same treatment. But the graduation from MIT would put that person above the graduation at SDSM.

Do they cover the same material and show the same level of proficiency? Apparantly not or you wouldn't choose a C student from MIT over a C student from SDSM.

As for 40% being a C, you need to look at foreign schools. It happens all the time. In many countries, getting above a 50% is a real accomplishment. Should these people be punished for being from a country where 50% is great? No. Instead, artifically convert the numbers into a relatively consistant system. Is it flawless? No. But at least you can say a person with a B average is a decent student - no matter what country they are from. You can't say the same about a student with a 60% average. Is that good or bad? Without knowing the country and detailed knowledge of the country's gradings system the 60% is meaningless.

OK, so if a student can get a C with a 40% at a foreign school, is that necessarily the same as a student who got a C with a 75% in the US? Different countries use different systems and have vastly different standards, I don't think it would ever be easy to compare grades from one country to another, nor do I think that should even be a factor in deciding what grading system to use in THIS country.

True, it doesn't fix everything. It fixes 90% of the problems, but on the gaps there are issues. But anytime you are on the border of a class, you should talk to the professor and get things straightened out. But if you really look at the whole scheme of things, a C+ vs a B- in one class has no significant impact on your GPA (we are talking about ~0.01 point difference in GPA). And no employer cares about 0.01 points.

Certainly, but I think given the choice between increased precision and decreased precision, I'd take the increased precision.

GPA is one of many statistics needed to compare people. Think of these two people, X and Y. Both are identical in every way. Except on one minor assignment on one class, person X got 6% taken off for a mistake. Person Y got 7% taken off for the same mistake. With GPAs, they both have the same class rank. With raw percentages, they have different class ranks with person X ahead of Y. GPAs do help in many of those situations.

Unless the difference of 1% on that assignment knocked the one guy from a B- to a C+. ;) Take the extreme case where one guy got an 83 in every class and another guy got an 80 in every class. Should they have the same class rank? :)
 

Savij

Diamond Member
Nov 12, 2001
4,233
0
71
It is because the percentage system is useless and unnecessary. The entire 0-60 range (almost always) stands for failure. Instead of using a large set of numbers that aren?t needed, use a small group of numbers that focuses on the useful section of the grading curve. The one through four scale focuses on the 60 (0 ? 0.99), 70 (1-1.99), 80(2-2.99), 90(3-3.99) range that is actually useful in grading.
 

dullard

Elite Member
May 21, 2001
26,066
4,712
126
Originally posted by: mugs
Unless the difference of 1% on that assignment knocked the one guy from a B- to a C+. ;) Take the extreme case where one guy got an 83 in every class and another guy got an 80 in every class. Should they have the same class rank? :)
I'm not fighting the rest any more. Like I said, the letters fix most of the issues. It wouldn't fix the B- to C+ border issue. But then, if one point puts you on the border, most professors are willing to talk about it. In my university, 83% was a B, 80% was a B-. Also my university gave a B a 3.0 rating and a B- a 2.666 rating. So in that case, they have dramatically different grades and GPAs.

 

mugs

Lifer
Apr 29, 2003
48,920
46
91
Originally posted by: dullard
Originally posted by: mugs
Unless the difference of 1% on that assignment knocked the one guy from a B- to a C+. ;) Take the extreme case where one guy got an 83 in every class and another guy got an 80 in every class. Should they have the same class rank? :)
I'm not fighting the rest any more. Like I said, the letters fix most of the issues. It wouldn't fix the B- to C+ border issue. But then, if one point puts you on the border, most professors are willing to talk about it. In my university, 83% was a B, 80% was a B-. Also my university gave a B a 3.0 rating and a B- a 2.666 rating. So in that case, they have dramatically different grades and GPAs.

OK, where the dividing line is doesn't matter... there's generally a 3 point range for each letter w/ + or -. So use 80 and 82 or 83 and 86, the point is the same... you're giving two people the same grade for levels of achievement that are far enough apart to matter. The fact that you would have to talk to a professor about a grade on the borderline, or that a professor would actually change your grade if it's on the borderline, makes the letter grading system all the more silly. Just do away with it and give the student exactly what they earned.
 

Colt45

Lifer
Apr 18, 2001
19,720
1
0
Originally posted by: Savij
It is because the percentage system is useless and unnecessary. The entire 0-60 range (almost always) stands for failure. Instead of using a large set of numbers that aren?t needed, use a small group of numbers that focuses on the useful section of the grading curve. The one through four scale focuses on the 60 (0 ? 0.99), 70 (1-1.99), 80(2-2.99), 90(3-3.99) range that is actually useful in grading.

but with a percentage thats not relevant, because you have enough accuracy to be able to span the whole scale.

gpa is useless and unecessary, considering its based on the percentage in the first place
 

dullard

Elite Member
May 21, 2001
26,066
4,712
126
Originally posted by: mugs
Just do away with it and give the student exactly what they earned.
That is just what letter grades are intended to do. Minor errors in grading or biases are ignored - so students get exactly what they earned. Fail, below average, average, above average, or excellent. That level of accuracy is basically all that is important for anything in the world. The fine details can cause enormous problems, so why keep it?

Your 83 vs 86 both are above average. Both would get the same jobs. Both would get into the same grad schools. Etc.
 

mugs

Lifer
Apr 29, 2003
48,920
46
91
Originally posted by: dullard
Originally posted by: mugs
Just do away with it and give the student exactly what they earned.
That is just what letter grades are intended to do. Minor errors in grading or biases are ignored - so students get exactly what they earned. Fail, below average, average, above average, or excellent. That level of accuracy is basically all that is important for anything in the world. The fine details can cause enormous problems, so why keep it?

Your 83 vs 86 both are above average. Both would get the same jobs. Both would get into the same grad schools. Etc.

When you're doing a large number of mathematical calculations, do you round after each calculation, or at the end? Same reasoning.
 

dullard

Elite Member
May 21, 2001
26,066
4,712
126
Originally posted by: mugs
When you're doing a large number of mathematical calculations, do you round after each calculation, or at the end? Same reasoning.
The whole intent of using letter grades is to round after each calculation. The intent is to smooth out differences and eliminate minor details. Grades are not an exact science, and thus they shouldn't be treated as one. They are approximate representations of your level of understanding. Using letters to iron out small details reflects the "approximate" part. That is the reason for them. I guess I can't get you to even see that side. Have a good day. And thanks for the good debate.

 

Savij

Diamond Member
Nov 12, 2001
4,233
0
71
Originally posted by: Colt45
Originally posted by: Savij
It is because the percentage system is useless and unnecessary. The entire 0-60 range (almost always) stands for failure. Instead of using a large set of numbers that aren?t needed, use a small group of numbers that focuses on the useful section of the grading curve. The one through four scale focuses on the 60 (0 ? 0.99), 70 (1-1.99), 80(2-2.99), 90(3-3.99) range that is actually useful in grading.

but with a percentage thats not relevant, because you have enough accuracy to be able to span the whole scale.

gpa is useless and unecessary, considering its based on the percentage in the first place

You know you're allowed to add extra decimal places to your GPA, right? You can list it as 1.99, 1.99, 1.999

Just like the percentages. Most of the time the percent is rounded to whole numbers, two digits. With the GPA it's rounded to the tenths, two digits. Typically. you have 40 numbers that are "useful" in the percentage system, and 40 numbers that are useful in the GPA system.