How average is "median"? Forum

[bdubs]

How big is the group at or around median at most schools? I have reviewed the curves for the law school that I will be attending (Northwestern) and understand the implementation of the curve in 1L classes. Median for a class should always be a B+. What I am less sure about it what the grade dispersion looks like across classes.

How many people will end up with a B+ average (or whatever the median grade is)?
Does it matter how you get to the GPA you have (consistency vs selective excellence)?

[usuaggie]

How big is the group at or around median at most schools? I have reviewed the curves for the law school that I will be attending (Northwestern) and understand the implementation of the curve in 1L classes. Median for the class should always be a B+. What I am less sure about it what the grade dispersion looks like across classes.

How many people will end up with a B+ average (or whatever the median grade is)?
Does it matter how you get to the GPA you have (consistency vs selective excellence)?

median is by definition standardized. i don't get what you are really asking. if the median is B+ 50% will be b+ or higher and 50% will be b+ or lower.

[bdubs]

How big is the group at or around median at most schools? I have reviewed the curves for the law school that I will be attending (Northwestern) and understand the implementation of the curve in 1L classes. Median for the class should always be a B+. What I am less sure about it what the grade dispersion looks like across classes.

How many people will end up with a B+ average (or whatever the median grade is)?
Does it matter how you get to the GPA you have (consistency vs selective excellence)?

median is by definition standardized. i don't get what you are really asking. if the median is B+ 50% will be b+ or higher and 50% will be b+ or lower.

I want to know the shape of the distribution. Are 50% of the people "close" to median, and 25% above/below? There could also be just a few people at median and a large number distributed across the high performing and low performing GPAs.

[somewhatwayward]

Does it matter how you get to the GPA you have (consistency vs selective excellence)?

the one thing that is good about inconsistency is, if you have performed very well in some areas, you can try to figure out what made you better in those areas as compared to your weaker classes. if you have all B+s, you don't have bright spots to extrapolate from.

[northwood]

I think OP is referring to the avctual number of students.
OP its pretty easy to figure out all you need is
number of 1L enrolled
Percentage or number of A. A- B+ etc grades
what the median grade is
You may have to hunt around on the schools website. Since its very school specific it may take a while if you have mor than 1 school in mind

[bdubs]

I think OP is referring to the avctual number of students.
OP its pretty easy to figure out all you need is
number of 1L enrolled
Percentage or number of A. A- B+ etc grades
what the median grade is
You may have to hunt around on the schools website. Since its very school specific it may take a while if you have mor than 1 school in mind

It may be that no student has all A+s, despite the fact that on average 5% of students in a given section get A+s. On the other hand it may be that those 5% A+s are always going to the same students so the GPA distribution looks a lot like the curve. That is what I want to know and as far as I know NU doesn't rank or publish this (at least in areas that I can access as an admit).

[sambeber]

EDIT: Apparently I don't understand the question.

[aliarrow]

I've wondered the same thing myself.

Sorry for all the excessively dickish responses that don't comprehend the question.
The tl;dr version of this thread so far is no one knows.

[bdubs]

You're going to Northwestern but can't figure out how to search for "northwestern law grades" on Google? Honestly.

This is the very first result.

http://www.law.northwestern.edu/academi ... olicy.html

Hope that helps.

No, that is what I already knew. How does no one understand the concept of a distribution?

Maybe pictures will help



OR



Obviously the shape of the distribution has to be influenced by the class distribution, but it is not determined by it.

[aliarrow]

If you want to be precise, I'd advise you ask for the standard deviation, as this is what you're looking for.

ETA: http://www.top-law-schools.com/forums/v ... 5&t=153322

[bdubs]

If you want to be precise, I'd advise you ask for the standard deviation, as this is what you're looking for.

Yes and no. If anyone can tell me what the standard deviation is that would be great, but i'm looking more for information on how people determine top XX% of your class when the shape of the distribution is not known. That involves knowing a bit more than the amount of variance in grades. At least you know what i'm talking about though.

[kalvano]

What relevance does this have?

It's not going to help you get a better grade.

[bdubs]

What relevance does this have?

It's not going to help you get a better grade.

Just for my own edification.

[kalvano]

What relevance does this have?

It's not going to help you get a better grade.

Just for my own edification.

Then what do you want to know? The exact number of people who get "x" grade? Or what? I don't think you can determine the exact amount without knowing how many people there, and that varies year to year.

[aliarrow]

Then what do you want to know? The exact number of people who get "x" grade? Or what? I don't think you can determine the exact amount without knowing how many people there, and that varies year to year.

It does have some real effect though. If there's a very normalized curve, then it would be easier to end up at median (though much more difficult to be top 20% or so, while also difficult to end up in the bottom 20%), if it's a flatter curve then the ranks are more volatile and risky.

[jkay]

Some schools divide up the class into quartiles or quintiles and post the ranges. For example, it might be something like

1st quintile: 4.0-3.65
2nd quintile: 3.64-3.4
3rd quintile: 3.2-.349
etc.

If so, you can obviously get a grasp of median GPA. If your school does not do this, there is still probably someone there with the institutional knowledge to help you figure it out.

But as a 0L, all of this is completely meaningless to you and you should not think about it at all.

[usuaggie]

the problem is your use of median. asking how many people are at median. there is only one point that is at the median. so you can't have 30 students at the median, unless they all got the exact same grades in the same credit-weighted classes.

[California Babe]

the problem is your use of median. asking how many people are at median. there is only one point that is at the median. so you can't have 30 students at the median, unless they all got the exact same grades in the same credit-weighted classes.

I believe OP's question is in regards to this problem:

Grade distribution 1: C, C, B-, B-, B, B, B, B, B, B, B, B, B, B, B, B, B, B+, B+, A, A

Grade distribution 2: C, C, C, C, C+, C+, B-, B-, B-, B-, B, B, B, B, B, B+, B+, B+, A-, A-, A,

Most people talk about "median" as being in the middle, with 50% above and 50% below. This is inaccurate (I realize you did not say this). In both of these scenarios, a B is a median grade, but in the first example only 4 students receive grades above or below a B, whereas in situation two 10 students receive below B grades, and only 6 students receive a grade below a B.

To comment on your claim that you can't have 30 students at median, that is false in the sense that if "B" is the median grade, you can have many (or few) students that are "at median." In the first example, 13/21 students are at median, whereas in the second situation, only 5/21 students are at median. So both grade distributions set B as a median, but a more students can say they are "at or above median" in the first situation than the second.

To answer OP's question: this depends on how your school enforces grading.

[bdubs]

Then what do you want to know? The exact number of people who get "x" grade? Or what? I don't think you can determine the exact amount without knowing how many people there, and that varies year to year.

It does have some real effect though. If there's a very normalized curve, then it would be easier to end up at median (though much more difficult to be top 20% or so, while also difficult to end up in the bottom 20%), if it's a flatter curve then the ranks are more volatile and risky.

Yes, I'd like to know the amount of stratification for employment purposes. If you have to be at median in order to get a shot at firm X, does 70% of the class have a shot, or just 51%. Same goes for any other %ile rank threshold, maybe 40% are actually at the GPA for top 1/3.

[Rurik]

the problem is your use of median. asking how many people are at median. there is only one point that is at the median. so you can't have 30 students at the median, unless they all got the exact same grades in the same credit-weighted classes.

To comment on your claim that you can't have 30 students at median, that is false in the sense that if "B" is the median grade, you can have many (or few) students that are "at median."

This is what he's saying, no?

[TaipeiMort]

the problem is your use of median. asking how many people are at median. there is only one point that is at the median. so you can't have 30 students at the median, unless they all got the exact same grades in the same credit-weighted classes.

I believe OP's question is in regards to this problem:

Grade distribution 1: C, C, B-, B-, B, B, B, B, B, B, B, B, B, B, B, B, B, B+, B+, A, A

Grade distribution 2: C, C, C, C, C+, C+, B-, B-, B-, B-, B, B, B, B, B, B+, B+, B+, A-, A-, A,

Most people talk about "median" as being in the middle, with 50% above and 50% below. This is inaccurate (I realize you did not say this). In both of these scenarios, a B is a median grade, but in the first example only 4 students receive grades above or below a B, whereas in situation two 10 students receive below B grades, and only 6 students receive a grade below a B.

To comment on your claim that you can't have 30 students at median, that is false in the sense that if "B" is the median grade, you can have many (or few) students that are "at median." In the first example, 13/21 students are at median, whereas in the second situation, only 5/21 students are at median. So both grade distributions set B as a median, but a more students can say they are "at or above median" in the first situation than the second.

To answer OP's question: this depends on how your school enforces grading.

Exactly. Median is also effected by how distributed the A grades are in a given class, which a grading scheme has no real control over.

Some years/schools the top 10% may be the same people in every class, some schools they may be completely dispersed. For example, at uChicago the "median" grade is a 177 (B), and top-ten percent at graduation is a 179.5, which is a B+, a top 5% is something like 180.5.

I would guess the distribution is something like:

C+,B-,B-(close to median),B,B,B,B,B,B+,A

this would mean that 80% are close to median or above.

[usuaggie]

the problem is your use of median. asking how many people are at median. there is only one point that is at the median. so you can't have 30 students at the median, unless they all got the exact same grades in the same credit-weighted classes.

To comment on your claim that you can't have 30 students at median, that is false in the sense that if "B" is the median grade, you can have many (or few) students that are "at median."

This is what he's saying, no?

that is exactly what i was saying.

[bdubs]

Some years/schools the top 10% may be the same people in every class, some schools they may be completely dispersed. For example, at uChicago the "median" grade is a 177 (B), and top-ten percent at graduation is a 179.5, which is a B+, a top 5% is something like 180.5.

I would guess the distribution is something like:

C+,B-,B-(close to median),B,B,B,B,B,B+,A

this would mean that 80% are close to median or above.

Thank you, this is the kind of thing I wanted to know. I realize it could vary year-to-year but I would imagine (perhaps falsely) that it doesn't.

[usuaggie]

at least at my school, teachers are told that the median grade, on a bell curve, needs to be between a b and a b- for first year courses. so it varies between teachers/sections. they just give papers the grades they think they deserve, and then tweak the grades to fit the curve if there are too many high or low. there are no set D's or A's they have to give or even want to give. If they want to, they can give everybody a B and that'd be fine

[Rurik]

OP, I'm guessing, it would be something like this:

3.4 = top 50%
3.5 = top third
3.58 = top quarter
3.63 = top 20%
3.67 = top 15%
3.74 = top 10%
3.8 = top 5%
3.88 = top 3%
3.95 = top 2%
4.00 = top 1%

Not sure if this is what you're looking for or not.