What percentage of ppl at T-14 at/above median?? Forum

[mallard]

Imagine a world where the OP isn't a moron,

Sorry, I'm gonna have to toke up for this.

and the question is "at any top 14 law school, what percentage of students will be viewed as roughly median or better by employers."

I'll grant that's NOT what OP asked, but it's pretty close - and it's an interesting and relevant question.

This is like basic statutory interpretation. If somebody says "how many are at or above median" you can either go "LOL 50% MORAN MEDIAN DICTIONARY HUR" and go play golf, or you can internalize the implication which is all but assuredly, as typed above, at any top 14 law school, what percentage of students will be viewed as roughly median or better by employers."

There is absolutely zero relationship between the OP's question about #52 out of 99 possibly having the same GPA as #50 and the issue of unpublished medians, man.

[rundoxierun]

Even if 60% was at median (which is probably beyond the upper bound possible), the 2/3 figure would just be raised to a number much closer to the median, and thus there would still be about 66 to 70% of people that are ranked in the top 2/3rd of the class....

A larger median does not influence the distribution of other percentile markings in a bell curve.

Im sorry, part of my question is kind of to what extent law school is on a bell curve?? if grades range from A=4.0 to C+=2.33 and a huge percentage of fall between A=4.0 and B+=3.33 then doesnt this skew it so it doesnt necessarily fit a bell curve. Obviously no one has access to this kind of data so question withdrawn.

[270910]

Imagine a world where the OP isn't a moron,

Sorry, I'm gonna have to toke up for this.

and the question is "at any top 14 law school, what percentage of students will be viewed as roughly median or better by employers."

I'll grant that's NOT what OP asked, but it's pretty close - and it's an interesting and relevant question.

This is like basic statutory interpretation. If somebody says "how many are at or above median" you can either go "LOL 50% MORAN MEDIAN DICTIONARY HUR" and go play golf, or you can internalize the implication which is all but assuredly, as typed above, at any top 14 law school, what percentage of students will be viewed as roughly median or better by employers."

There is absolutely zero relationship between the OP's question about #52 out of 99 possibly having the same GPA as #50 and the issue of unpublished medians, man.

Look, even if the median is published.

Take my school again. It, like the majority of top tierish law schools, has a "3.3 median". First, it sneaks just a bit above a 3.3 by the time all the classes are done and generosity is given.

So I can confidantly say the following things about my school:

1) The median GPA in every class is roughly 3.3

2) The median GPA for the student body is roughly 3.3

3) Our transcripts proudly proclaim that the median GPA given out is roughly 3.3

4) The number of students who have a gpa at 3.30 or higher is going to be noticeably above 50%. I can't speculate as to precisely how much higer, but it will be higher.

Relevant to anything? Nope. But did I just type it all out? Because I think I did.

Wait, what?

WAIT WHO.

...are you my mommy?

[rundoxierun]

Yes, I agree. But it is implicitly, halfway. Example:
Class is: 3.8, 3.6, 3.6, 3.5, 3.4, 3.3, 3.2, 2.9
Median is: 3.5, or could be an specific number between 3.5 and 3.4 but those numbers exactly. I.e. median could be 3.41 (assuming the above GPAs were not rounded.)
Mode is: 3.6

But the median is calculated as the value at which half the set is above and half the set is below.

Ok, I see what you are getting at. I shouldve said at/or above the gpa of the median student.

[Cavalier]

...are you my mommy?

Yes.

[Na_Swatch]

Even if 60% was at median (which is probably beyond the upper bound possible), the 2/3 figure would just be raised to a number much closer to the median, and thus there would still be about 66 to 70% of people that are ranked in the top 2/3rd of the class....

A larger median does not influence the distribution of other percentile markings in a bell curve.

Im sorry, part of my question is kind of to what extent law school is on a bell curve?? if grades range from A=4.0 to C+=2.33 and a huge percentage of fall between A=4.0 and B+=3.33 then doesnt this skew it so it doesnt necessarily fit a bell curve. Obviously no one has access to this kind of data so question withdrawn.

Ok, much better question. Many schools use a system that is pretty close to normal distribution (thus approximating a bell curve). The fact is for 4 to 5 classes each semester and grades ranging from A to C+, the vast majority of schools will have enough difference between final GPA's that the Median calculation will almost always result in about 50% (maybe + an additional couple % points) falling at or above.

However, if you have only H, P, LP grades then the possibility for a larger amount at Median becomes much more likely.

[Na_Swatch]

Look, even if the median is published.

Take my school again. It, like the majority of top tierish law schools, has a "3.3 median". First, it sneaks just a bit above a 3.3 by the time all the classes are done and generosity is given.

So I can confidantly say the following things about my school:

1) The median GPA in every class is roughly 3.3

2) The median GPA for the student body is roughly 3.3

3) Our transcripts proudly proclaim that the median GPA given out is roughly 3.3

4) The number of students who have a gpa at 3.30 or higher is going to be noticeably above 50%. I can't speculate as to precisely how much higer, but it will be higher.

Relevant to anything? Nope. But did I just type it all out? Because I think I did.

Wait, what?

WAIT WHO.

...are you my mommy?

This is obsfucation then and some students claiming to be at or "above median" are fudging the numbers.

The school is probably helping this along by stating that the Median is 3.3 when it should be 3.3x

[mec30]

Even if 60% was at median (which is probably beyond the upper bound possible), the 2/3 figure would just be raised to a number much closer to the median, and thus there would still be about 66 to 70% of people that are ranked in the top 2/3rd of the class....

A larger median does not influence the distribution of other percentile markings in a bell curve.

Im sorry, part of my question is kind of to what extent law school is on a bell curve?? if grades range from A=4.0 to C+=2.33 and a huge percentage of fall between A=4.0 and B+=3.33 then doesnt this skew it so it doesnt necessarily fit a bell curve. Obviously no one has access to this kind of data so question withdrawn.

To curve a class means to adjust it to fit a bell curve. Assuming there are no outliers, a graphed bell curve would consider half of those GPAs above the highpoint and half below, regardless of the fact that those 10 at median students (for example) are indistinguishable from each other in terms of GPA.

[rundoxierun]

Yes, I agree. But it is implicitly, halfway. Example:
Class is: 3.8, 3.6, 3.6, 3.5, 3.4, 3.3, 3.2, 2.9
Median is: 3.5, or could be an specific number between 3.5 and 3.4 but those numbers exactly. I.e. median could be 3.41 (assuming the above GPAs were not rounded.)
Mode is: 3.6

But the median is calculated as the value at which half the set is above and half the set is below.

Ok, I see what you are getting at. I shouldve said at/or above the gpa of the median student.

What? That doesn't change anything? WTF is going on here? Am I getting trolled?

If 201 gpas are listed out from highest to lowest and someone counted 101 of them from either end how many gpas would have a value equal to or higher than that 101st gpa. Median in statistical terms was a bad way to put it.

[270910]

Imagine a world where the OP isn't a moron,

Sorry, I'm gonna have to toke up for this.

and the question is "at any top 14 law school, what percentage of students will be viewed as roughly median or better by employers."

I'll grant that's NOT what OP asked, but it's pretty close - and it's an interesting and relevant question.

This is like basic statutory interpretation. If somebody says "how many are at or above median" you can either go "LOL 50% MORAN MEDIAN DICTIONARY HUR" and go play golf, or you can internalize the implication which is all but assuredly, as typed above, at any top 14 law school, what percentage of students will be viewed as roughly median or better by employers."

There is absolutely zero relationship between the OP's question about #52 out of 99 possibly having the same GPA as #50 and the issue of unpublished medians, man.

Look, even if the median is published.

Take my school again. It, like the majority of top tierish law schools, has a "3.3 median". First, it sneaks just a bit above a 3.3 by the time all the classes are done and generosity is given.

So I can confidantly say the following things about my school:

1) The median GPA in every class is roughly 3.3

2) The median GPA for the student body is roughly 3.3

3) Our transcripts proudly proclaim that the median GPA given out is roughly 3.3

4) The number of students who have a gpa at 3.30 or higher is going to be noticeably above 50%. I can't speculate as to precisely how much higer, but it will be higher.

Relevant to anything? Nope. But did I just type it all out? Because I think I did.

Wait, what?

WAIT WHO.

...are you my mommy?

To be fair, you just said in that statement that the median was roughly 3.3, not exactly 3.3.

You're entirely correct, as is the other guy.

The point is to contextualize GPA. At top schools, firms don't get your rank, they get your GPA and the school saying "lololol 3.3 median lololol hire our grads."

I'm really not trying to argue that it is especially relevant, just that as an interesting and perhaps intentional quirk more students are going to be viewed as 50th percentile than are 50th percentile.

You keep arguing about the mathematical definition of median and I concede that and have never argued that. Yes. Truth. Accurate. But my point (which isn't a very interesting or relevant point, so I don't know why I'm typing) is that you have to view grades through the lens of their actual effect when viewed by employers. In that respect, having a 3.3 instead of a 3.29 or whatever might be relevant psychologically.

Marginally, I know. And if I'm answering a question it's one nobody asked. Yes yes I get it. But my point is we're saying two different things. At least. Maybe more. Huh. What?

Who?

Meta.

[rundoxierun]

You're entirely correct, as is the other guy.

The point is to contextualize GPA. At top schools, firms don't get your rank, they get your GPA and the school saying "lololol 3.3 median lololol hire our grads."

I'm really not trying to argue that it is especially relevant, just that as an interesting and perhaps intentional quirk more students are going to be viewed as 50th percentile than are 50th percentile.

You keep arguing about the mathematical definition of median and I concede that and have never argued that. Yes. Truth. Accurate. But my point (which isn't a very interesting or relevant point, so I don't know why I'm typing) is that you have to view grades through the lens of their actual effect when viewed by employers. In that respect, having a 3.3 instead of a 3.29 or whatever might be relevant psychologically.

Marginally, I know. And if I'm answering a question it's one nobody asked. Yes yes I get it. But my point is we're saying two different things. At least. Maybe more. Huh. What?

Who?

Meta.

Thanks very much. I am in fact a 0L and just wanted to ask a question on how this worked. I wasnt aware that at most top schools employers do not have access to accurate data on class rank(thought this situation was the exception). Just trying to get a handle on how these things shake out. Basically, prepping for the whole law school app thing and will be pretty obsessed about it til october 1stish so asking questions to occupy my time at work.. ppl really need to chill out.

[mallard]

Thanks very much. I am in fact a 0L and just wanted to ask a question on how this worked. I wasnt aware that at most top schools employers do not have access to accurate data on class rank(thought this situation was the exception). Just trying to get a handle on how these things shake out. Basically, prepping for the whole law school app thing and will be pretty obsessed about it til october 1stish so asking questions to occupy my time at work.. ppl really need to chill out.

This is not worthwhile to think about even as a law student.

[09042014]

Disco_Barred's first semester courses and the number of grades given out in each that were at or above median:

Course 1: 74.9%
Course 2: 59.2%
Course 3: 64.6%
Course 4: 63.1%

1) With respect to individual classes: At a school with a B+ curve, well over half of each class gets median or better. Often 60, 70%+.

The median grade would then be higher than the median grade that each course is curved to.

This is just flat wrong, you're thinking about it the wrong way. Each of those courses had a 3.3 median GPA (roughly), the school had a 3.3 median GPA (roughly).

This is highly unlikely unless that bottom 40% in each class were remarkably consistently the bottom 40%.

[rundoxierun]

Thanks very much. I am in fact a 0L and just wanted to ask a question on how this worked. I wasnt aware that at most top schools employers do not have access to accurate data on class rank(thought this situation was the exception). Just trying to get a handle on how these things shake out. Basically, prepping for the whole law school app thing and will be pretty obsessed about it til october 1stish so asking questions to occupy my time at work.. ppl really need to chill out.

This is not worthwhile to think about even as a law student.

Im a hyper-rational type that thinks about the world as if it was an econ scenario. I like to collect information on all kinds of random stuff. Sorry if I offended the TLS community by asking a simple question regarding the percentage of people who have equivalent gpas or grades that would be seen as the median.

[270910]

The median grade would then be higher than the median grade that each course is curved to.

This is just flat wrong, you're thinking about it the wrong way. Each of those courses had a 3.3 median GPA (roughly), the school had a 3.3 median GPA (roughly).

This is highly unlikely unless that bottom 40% in each class were remarkably consistently the bottom 40%.

Disco_barred: Went to this law school, looked at the data from this law school, typed it into a little box and pressed submit

Desert Fox: Calls a fact (with no interpretation from me) "unlikely"

i don't meant to be too snarky. Maybe it will help visualize to show you how the curve worked:

The 75% class has the neatest numbers:

A: 5%
A-: 20%
B+: 50%
B: 20%
B-: 5%

See how that works? Nice, neat curve - median 3.3 - but 75% of the class is getting a grade which is "median or better." Not unrealistic or unlikely or imagination stretching at all. It's actually just simple and symmetrical.

[romothesavior]

Yeah, the "LOL MORAN" brigade was really out in unnecessary force on this one.

This,

[romothesavior]

I just did a whole first years worth of classes at Michigan -- I was generous with the curve (because it moves) to put more people AT or ABOVE median and got 54%

http://mblog.lib.umich.edu/RG/archives/ ... index.html

have at it. 1st year classes only, though.

This was the type of info OP was looking for, and should have been your response in the first place.

And also this.

[rundoxierun]

The school would then say the median was rough or approximate, because there would be one student above or below in excess of 50%.

You are losing me again.. that 101st gpa is not necessarily a unique value.. the next 10 gpas on the list could be the exact same. This would mean that 111/201 students had a gpa at or above that 101st student's gpa.

[mec30]

Guys, you're being really dumb. His question is legitimate.

1) With respect to individual classes: At a school with a B+ curve, well over half of each class gets median or better. Often 60, 70%+.

2) With respect to the performance of the class as a whole, it's still quite possible that more than half of the class is median or better. The normal distribution suggests that the most populous region of the curve after 1L year is likely to be at or within a few points of the median. If you take a liberal definition of 'median' extending to just below it - or if you are at a school with an amorphous 3.15-3.25 type median instead of one occurring right at a letter (such as 3.3, or 178 or however the hell Chicago does it) - there will actually be a sizable enough portion of the population directly 'at median' that it would be accurate to say more than half of the class is at the median. In sum, my guess is up to 60% of any given class is "at or above median" by a definition most reasonable people would agree on for "median."

But by all means, get back to trolling the shit out of the OP -.-

Just no. Look, the class is either going to have an odd number of students or an even number of students. If there are an even number of students, then the school will pick the lowest student in the top half, and that will be above median. The rest will be below median, and the division will be equal. If, on the other hand, there are an odd number of students, then it is true that the division will not be exactly 50/50. But, it will only be off by 1 student. That's just the way a median works. Median is a statistically significant term. If you are talking about how many people are at or above the curve, sure there may be some play there—but not with a median figure.

Median is based on values not datapoints.. the median is the halfway VALUE of population not datapoint(that would be mode).

Yes, I agree. But it is implicitly, halfway. Example:
Class is: 3.8, 3.6, 3.6, 3.5, 3.4, 3.3, 3.2, 2.9
Median is: 3.5, or could be an specific number between 3.5 and 3.4 but those numbers exactly. I.e. median could be 3.41 (assuming the above GPAs were not rounded.)
Mode is: 3.6

But the median is calculated as the value at which half the set is above and half the set is below.

How did you get 3.41 as the median? You can't add the two ends together and divide.

[sbalive]

Obviously no particular class is graded on an actual bell "curve," given that letter grades are discrete values. But, an entire section or all students in a given year, there will be something very close to an actually continuous "bell curve" of GPA's for all the students, pretty much no matter whether the distribution of grades in each class actually is symmetric or if 73% of the students in each individual class get B+'s or whatever.

[sbalive]

Yes, I agree. But it is implicitly, halfway. Example:
Class is: 3.8, 3.6, 3.6, 3.5, 3.4, 3.3, 3.2, 2.9
Median is: 3.5, or could be an specific number between 3.5 and 3.4 but those numbers exactly. I.e. median could be 3.41 (assuming the above GPAs were not rounded.)
Mode is: 3.6

But the median is calculated as the value at which half the set is above and half the set is below.

How did you get 3.41 as the median? You can't add the two ends together and divide.

Sigh.

[mec30]

Yes, I agree. But it is implicitly, halfway. Example:
Class is: 3.8, 3.6, 3.6, 3.5, 3.4, 3.3, 3.2, 2.9
Median is: 3.5, or could be an specific number between 3.5 and 3.4 but those numbers exactly. I.e. median could be 3.41 (assuming the above GPAs were not rounded.)
Mode is: 3.6

But the median is calculated as the value at which half the set is above and half the set is below.

How did you get 3.41 as the median? You can't add the two ends together and divide.

Sigh.

If the last GPA was 2.7, instead of 2.9, the calulated median (using them method above) would be 3.39x and that would be completly wrong.

Neither 3.5 nor 3.41 is the median.

[rundoxierun]

Yes, I agree. But it is implicitly, halfway. Example:
Class is: 3.8, 3.6, 3.6, 3.5, 3.4, 3.3, 3.2, 2.9
Median is: 3.5, or could be an specific number between 3.5 and 3.4 but those numbers exactly. I.e. median could be 3.41 (assuming the above GPAs were not rounded.)
Mode is: 3.6

But the median is calculated as the value at which half the set is above and half the set is below.

How did you get 3.41 as the median? You can't add the two ends together and divide.

Sigh.

If the last GPA was 2.7, instead of 2.9, the calulated median (using them method above) would be 3.39x and that would be completly wrong.

I think the "I.e. 3.41" was just referring to a random datapoint between 3.4 and 3.5, not actual median calculation. Regardless, the example doesnt really relate to my question. My question specifically dealt with situations where the calculated median would not be a unique value(i.e. the calculated median is 3.4 and a lot of people have 3.4 gpas.).

[mec30]

How did you get 3.41 as the median? You can't add the two ends together and divide.

Sigh.

If the last GPA was 2.7, instead of 2.9, the calulated median (using them method above) would be 3.39x and that would be completly wrong.

I think the "I.e. 3.41" was just referring to a random datapoint between 3.4 and 3.5, not actual median calculation. Regardless, the example doesnt really relate to my question. My question specifically dealt with situations where the calculated median would not be a unique value(i.e. the calculated median is 3.4 and a lot of people have 3.4 gpas.).

The problem is that’s not how medians are calculated. Traditionally, if there is an even number of terms the median is the mean of the two middle numbers. In which case neither 3.5 nor 3.41 is the median of that set.

[mec30]

And it's relevent to your question because if you changed the data set to {3.8, 3.6, 3.3, 3.3, 3.3, 3.2, 3.1, 2.9} using the way he calulated the last problem you would end up with a median at 3.31x instead of 3.3.