Curve calculation help

(Study Tips, Dealing With Stress, Maintaining a Social Life, Financial Aid, Internships, Bar Exam, Careers in Law . . . )
User avatar
zanda
Posts: 526
Joined: Fri Apr 24, 2009 9:36 am

Re: Curve calculation help

Postby zanda » Sun Jan 24, 2010 8:44 pm

Splitt3r wrote:
rayiner wrote:
drew wrote:seal, can you figure out anything from GULC's grade curve as to the median GPA, etc. (realizing that its completely speculative)? Its:

A 12%

A- 19%

B+ 28%

B 31%-36%

B- and below 5%-10%


Median is a B+, mean is probably around 3.3-3.33.


Drew and I were talking about this earlier, and though I see why it seems tempting to say that the median is a 3.3 exactly, I think at this point (we have 3 classes/11 units of grades) it's probably a 3.2 or so. I imagine that the As are pretty tightly grouped or at least that it's more often than not true that a person who got one A got two or more As, which would mean that we'd have a fair number of 3.7-3.8ish gpas, but relatively few 3.5s. Given the large percentage of Bs they give out, and the fact that B+/B line is relatively close to the natural median, it seems like the most likely scenario for the median person's grade distribution would be B+, B+, B. Obviously this will balance out over time and go towards a 3.3, but after just the first semester I think it's likely the case.

this is a similar issue that I've had with the NYU curve

User avatar
Nazrix
Posts: 47
Joined: Wed Dec 27, 2006 12:05 pm

Re: Curve calculation help

Postby Nazrix » Sat Jan 30, 2010 7:49 am

As a math major from one of the best public undergrads in the country, I can assure you, your statistics are based on the wildest of theories.

User avatar
Nazrix
Posts: 47
Joined: Wed Dec 27, 2006 12:05 pm

Re: Curve calculation help

Postby Nazrix » Sat Jan 30, 2010 8:49 am

thesealocust wrote:
blzrchick2 wrote:Hey TheSeaLocust, here's another one for you, if you have the time!

Median: 3.047
Top 33%: 3.275
Top 25%: 3.400
Top 10%: 3.615

What rough percentile is a 3.739?

Thanks!!


:roll:

There's no point in doing the math once you're above the top 10%. If it's by a noticeable margin, you're probably top 5%. But until you have a lot of data points, you can't actually use normal curve behavior to estimate. By definition if you have exceeded the top 10% by a substantial margin, there aren't going to be enough people (read: enough data) for the calculation to be meaningful, because outliers are (by their nature) unpredictable. One year there may be 5 4.0s and nobody between 3.8 and 4.0, the next year there might be a huge clump of people at 3.7 and you'd be top of the class. You can't answer those questions with statistics after one semester of grades.

That being said, I can't stop myself from doing the math, which spits out the number ~top 6%.


Previous post where I claimed I am a math major and your theories are ridiculous directed specifically at this comment, but applies generally to your other speculations. First, stop saying "you ran the math," you are either shorthand guessing based on part-math and some "guess" that's not a stats formula, or you are plugging data into a program...now I don't have problem with your (computer's) guessing, but I have a problem with your arrogant analysis of what the computer told you...

Do you even know what a curve is? Can you truly assume that the top 10%of the class "outliers" are really grabbing all the top grades in a limited curve? For 3 years? And it's going to vary some 3 year-sets from a 4.0 to a 3.7? What the hell?

Assuming it's some semblance of a standard curve, where 10% of a given grade set in a given course gets a 4.0, you are now assuming so many in the top 10%+ of the entire class, will be taking most/all of the top 10% in a given course? Rather than just having a reasonably good average over all their classes?

I don't know where you are attributing this uniqueness among top 10, but it is certainly not statistical study. This is not an income distribution model, we have a known high-point and a known distribution of grades in a given course and how many there actually can be numerically to get near the ceiling...

Statistics and basic math demonstrates that in this situation it becomes significantly HARDER for the top 10% (read: less likely) to be outliers from blzrchick2's stats on the %tiles, with so few grades that can pull one up from a 3.6x to a 4.0, in an given course (read: so many less even "chances" for a given person to successively pull those very few grades.)

With such a limited ceiling available to pull grades up dramatically, for a person on that curve, and the concept that ones GPA is composed of several different ranges, the upper range is very stunted, which would make more sense that people would clump much closer to the 3.615 top 10%, not away from it into the upper echelons of nonsense-theory you make up to trick law students who can't calculate a simple tip at a restaurant...

User avatar
thesealocust
Posts: 8448
Joined: Mon Oct 20, 2008 8:50 pm

Re: Curve calculation help

Postby thesealocust » Sat Jan 30, 2010 11:12 am

edit: never mind
Last edited by thesealocust on Wed Jun 30, 2010 11:07 pm, edited 1 time in total.

User avatar
Nazrix
Posts: 47
Joined: Wed Dec 27, 2006 12:05 pm

Re: Curve calculation help

Postby Nazrix » Sat Jan 30, 2010 5:58 pm

thesealocust wrote:
Nazrix wrote:
thesealocust wrote:
blzrchick2 wrote:Hey TheSeaLocust, here's another one for you, if you have the time!

Median: 3.047
Top 33%: 3.275
Top 25%: 3.400
Top 10%: 3.615

What rough percentile is a 3.739?

Thanks!!


:roll:

There's no point in doing the math once you're above the top 10%. If it's by a noticeable margin, you're probably top 5%. But until you have a lot of data points, you can't actually use normal curve behavior to estimate. By definition if you have exceeded the top 10% by a substantial margin, there aren't going to be enough people (read: enough data) for the calculation to be meaningful, because outliers are (by their nature) unpredictable. One year there may be 5 4.0s and nobody between 3.8 and 4.0, the next year there might be a huge clump of people at 3.7 and you'd be top of the class. You can't answer those questions with statistics after one semester of grades.

That being said, I can't stop myself from doing the math, which spits out the number ~top 6%.


Previous post where I claimed I am a math major and your theories are ridiculous directed specifically at this comment, but applies generally to your other speculations. First, stop saying "you ran the math," you are either shorthand guessing based on part-math and some "guess" that's not a stats formula, or you are plugging data into a program...now I don't have problem with your (computer's) guessing, but I have a problem with your arrogant analysis of what the computer told you...

Do you even know what a curve is? Can you truly assume that the top 10%of the class "outliers" are really grabbing all the top grades in a limited curve? For 3 years? And it's going to vary some 3 year-sets from a 4.0 to a 3.7? What the hell?

Assuming it's some semblance of a standard curve, where 10% of a given grade set in a given course gets a 4.0, you are now assuming so many in the top 10%+ of the entire class, will be taking most/all of the top 10% in a given course? Rather than just having a reasonably good average over all their classes?

I don't know where you are attributing this uniqueness among top 10, but it is certainly not statistical study. This is not an income distribution model, we have a known high-point and a known distribution of grades in a given course and how many there actually can be numerically to get near the ceiling...

Statistics and basic math demonstrates that in this situation it becomes significantly HARDER for the top 10% (read: less likely) to be outliers from blzrchick2's stats on the %tiles, with so few grades that can pull one up from a 3.6x to a 4.0, in an given course (read: so many less even "chances" for a given person to successively pull those very few grades.)

With such a limited ceiling available to pull grades up dramatically, for a person on that curve, and the concept that ones GPA is composed of several different ranges, the upper range is very stunted, which would make more sense that people would clump much closer to the 3.615 top 10%, not away from it into the upper echelons of nonsense-theory you make up to trick law students who can't calculate a simple tip at a restaurant...


:lol:


gunner.

thegrandinquisitor
Posts: 23
Joined: Sun Mar 15, 2009 10:33 pm

Re: Curve calculation help

Postby thegrandinquisitor » Sat Jan 30, 2010 6:53 pm

.
Last edited by thegrandinquisitor on Fri Apr 30, 2010 3:12 pm, edited 1 time in total.

User avatar
thesealocust
Posts: 8448
Joined: Mon Oct 20, 2008 8:50 pm

Re: Curve calculation help

Postby thesealocust » Sat Jan 30, 2010 7:05 pm

edit: never mind
Last edited by thesealocust on Wed Jun 30, 2010 11:07 pm, edited 1 time in total.

absworkoutplan
Posts: 24
Joined: Wed Oct 29, 2008 1:05 pm

Re: Curve calculation help

Postby absworkoutplan » Sun Jan 31, 2010 5:02 am

I figure someone might be able to do my school. They release grade distributions, but nothing on GPA so I only have a rough idea. The curve was also slightly looser (about 2-3%) looser this semester than last year. This breakdown only applies to my section, but I figure the other two were similar enough.

A : 14-17% (average 16%)
A- : 14-17% (average 16%)
B+ : 30-31% (usually 30%)
B : 30-31% (usually 31%)
B- : 7-9% (usually 8%)

A+'s and C's are discretionary and there were perhaps 2 of each in my section, so I don't think they really matter. Top 62% seemed to be B+ and above, which is higher than the usual 58-59%. I think median would probably be around a 3.33, but i'm not sure. I also hit about a 3.49 so if anyone has any idea where that would be, I would appreciate it.

User avatar
thesealocust
Posts: 8448
Joined: Mon Oct 20, 2008 8:50 pm

Re: Curve calculation help

Postby thesealocust » Sun Jan 31, 2010 11:09 am

edit: never mind
Last edited by thesealocust on Wed Jun 30, 2010 11:07 pm, edited 1 time in total.

User avatar
rayiner
Posts: 6184
Joined: Thu Dec 11, 2008 11:43 am

Re: Curve calculation help

Postby rayiner » Sun Jan 31, 2010 11:20 am

thesealocust wrote:
absworkoutplan wrote:I figure someone might be able to do my school. They release grade distributions, but nothing on GPA so I only have a rough idea. The curve was also slightly looser (about 2-3%) looser this semester than last year. This breakdown only applies to my section, but I figure the other two were similar enough.

A : 14-17% (average 16%)
A- : 14-17% (average 16%)
B+ : 30-31% (usually 30%)
B : 30-31% (usually 31%)
B- : 7-9% (usually 8%)

A+'s and C's are discretionary and there were perhaps 2 of each in my section, so I don't think they really matter. Top 62% seemed to be B+ and above, which is higher than the usual 58-59%. I think median would probably be around a 3.33, but i'm not sure. I also hit about a 3.49 so if anyone has any idea where that would be, I would appreciate it.


Based on other schools. probably top third to top 20%, but there's nothing you can do with the distributions for each class, you have to use a data point about the class as a whole (i.e. top 10% of the class is X GPA) because the individual classes will have different curves than the group behavior. Still, I'd wager large sums of money that your rank is in the top 20-30% range based on having done a lot of these calculations in the past.


I think that's doubtful. If the A+ counts as 4.33 and a couple are actually given out, then this curve seems slightly more lenient than the NU curve (more grades at/above B+, fewer grades below). 3.5 is probably outside the top 1/3 on our 1L curve.

User avatar
thesealocust
Posts: 8448
Joined: Mon Oct 20, 2008 8:50 pm

Re: Curve calculation help

Postby thesealocust » Sun Jan 31, 2010 11:30 am

edit: never mind
Last edited by thesealocust on Wed Jun 30, 2010 11:06 pm, edited 1 time in total.

User avatar
rayiner
Posts: 6184
Joined: Thu Dec 11, 2008 11:43 am

Re: Curve calculation help

Postby rayiner » Sun Jan 31, 2010 12:53 pm

thesealocust wrote:
rayiner wrote:I think that's doubtful. If the A+ counts as 4.33 and a couple are actually given out, then this curve seems slightly more lenient than the NU curve (more grades at/above B+, fewer grades below). 3.5 is probably outside the top 1/3 on our 1L curve.


Nothing on God's earth is more lenient than the NU curve. Nothing I say! Your professors HAVE to give 3% A+s, and can give up to 7% - which is all but certainly going to be more than a discretionary grant at the end of the day. Then your profs can give 25-37(!!!) percent 'A range' grades.


My impression is that professors do not give out the maximum number of A+ grades they're allowed to. The mandatory A+'s amount to two in a 62-person section, which is the same as what the OP mentioned were discretionarily awarded in his section. As for the rest of the curve:

OP's school:
A : 14-17% (average 16%)
A- : 14-17% (average 16%)
B+ : 30-31% (usually 30%)
B : 30-31% (usually 31%)
B- : 7-9% (usually 8%)

NU (sample from real 1L class, posted by chadwick):
A+: 3-7% (3%)
A : 12-15% (15%)
A-: 10-15% (15%)
B+: 15-30% (29%)
B: 20-35% (27%)
B-: 10-15% (11%)

VMan
Posts: 3
Joined: Mon Oct 12, 2009 2:03 am

Re: Curve calculation help

Postby VMan » Sun Jan 31, 2010 2:35 pm

thegrandinquisitor wrote:Sorry, can somebody please do it for my school.

Median is probably a 3.2

Top 20% is a 3.528

What % is a 3.728?


What would the BOTTOM 10-20% be for this school? (Which is almost certainly the school I attend, and where I failed pret-ty hard last semester!)

lnt
Posts: 2
Joined: Wed Aug 05, 2009 3:00 pm

Re: Curve calculation help

Postby lnt » Sun Jan 31, 2010 2:50 pm

10% 3.57
25% 3.38
33% 3.31
50% 3.19

What's top 5%/3% - i'm guessing around 3.7/3.8?

Tyvm

absworkoutplan
Posts: 24
Joined: Wed Oct 29, 2008 1:05 pm

Re: Curve calculation help

Postby absworkoutplan » Sun Jan 31, 2010 4:58 pm

rayiner wrote:
thesealocust wrote:
rayiner wrote:I think that's doubtful. If the A+ counts as 4.33 and a couple are actually given out, then this curve seems slightly more lenient than the NU curve (more grades at/above B+, fewer grades below). 3.5 is probably outside the top 1/3 on our 1L curve.


Nothing on God's earth is more lenient than the NU curve. Nothing I say! Your professors HAVE to give 3% A+s, and can give up to 7% - which is all but certainly going to be more than a discretionary grant at the end of the day. Then your profs can give 25-37(!!!) percent 'A range' grades.


My impression is that professors do not give out the maximum number of A+ grades they're allowed to. The mandatory A+'s amount to two in a 62-person section, which is the same as what the OP mentioned were discretionarily awarded in his section. As for the rest of the curve:

OP's school:
A : 14-17% (average 16%)
A- : 14-17% (average 16%)
B+ : 30-31% (usually 30%)
B : 30-31% (usually 31%)
B- : 7-9% (usually 8%)

NU (sample from real 1L class, posted by chadwick):
A+: 3-7% (3%)
A : 12-15% (15%)
A-: 10-15% (15%)
B+: 15-30% (29%)
B: 20-35% (27%)
B-: 10-15% (11%)


Just to qualify my statement and see if it affects your data here, that was amongst all 4 of our substantive classes combined, not in each class. Meaning 2 professors decided to give out 1 A+ each, and one gave out two C's. Not sure if there was confusion regarding that. Also our sections are of 80-something, so I think that comes out to less than a 0.5% chance of getting an A+ in any given class. But yeah, our B+ and higher distribution this year is not the norm as I said, after having looked at the last 2 years.

lnt
Posts: 2
Joined: Wed Aug 05, 2009 3:00 pm

Re: Curve calculation help

Postby lnt » Tue Feb 02, 2010 3:03 pm

lnt wrote:10% 3.57
25% 3.38
33% 3.31
50% 3.19

What's top 5%/3% - i'm guessing around 3.7/3.8?

Tyvm

Please?!

VMan
Posts: 3
Joined: Mon Oct 12, 2009 2:03 am

Re: Curve calculation help

Postby VMan » Tue Feb 02, 2010 5:10 pm

VMan wrote:
thegrandinquisitor wrote:Sorry, can somebody please do it for my school.

Median is probably a 3.2

Top 20% is a 3.528

What % is a 3.728?


What would the BOTTOM 10-20% be for this school? (Which is almost certainly the school I attend, and where I failed pret-ty hard last semester!)


P.S. I realize now that, since it is a normal curve, the answer to this would just be the inverse of the top 20%, essentially (i.e. 0.328 BELOW 3.2 rather than 0.328 above it)...You can see the sort of intellectual engagement that led me to this portion of my class.

afterglow99
Posts: 71
Joined: Sun Oct 26, 2008 12:24 am

Re: Curve calculation help

Postby afterglow99 » Tue Feb 02, 2010 6:46 pm

Here's a curveball - My school doesn't rank and I'm trying to figure out my approximate class rank anyway. I know my gpa falls exactly at an A- (3.67) and I know the school has a mandatory median set between 2.9 and 3.1.

I've acquired the grade distributions for the classes I took. If I can get access to the distributions for the other 1L sections, I would then be able to see the absolute lowest percentile rank that my gpa falls at, correct? (by looking at the number of As given out in each class and assuming that the same people who got one A got all As).

Is this off base?

User avatar
badfish
Posts: 1160
Joined: Mon Jun 02, 2008 12:53 pm

Re: Curve calculation help

Postby badfish » Tue Feb 02, 2010 7:04 pm

afterglow99 wrote:Here's a curveball - My school doesn't rank and I'm trying to figure out my approximate class rank anyway. I know my gpa falls exactly at an A- (3.67) and I know the school has a mandatory median set between 2.9 and 3.1.

I've acquired the grade distributions for the classes I took. If I can get access to the distributions for the other 1L sections, I would then be able to see the absolute lowest percentile rank that my gpa falls at, correct? (by looking at the number of As given out in each class and assuming that the same people who got one A got all As).

Is this off base?


Your life is off base. You have a 3.67, you need not worry about the rest.




Return to “Forum for Law School Students”

Who is online

Users browsing this forum: pkou19 and 3 guests