Top Law Schools

No idea what I am doing. Fuck.

bjsesq wrote:No idea what I am doing. Fuck.

Someone broke it again. Probably you.

existenz wrote:

bjsesq wrote:No idea what I am doing. Fuck.

Someone broke it again. Probably you.

I stopped immediately once I realized that something wasn't right, but yeah, I probably fucked it up.

bjsesq wrote:

existenz wrote:

bjsesq wrote:No idea what I am doing. Fuck.

Someone broke it again. Probably you.

I stopped immediately once I realized that something wasn't right, but yeah, I probably fucked it up.

Fixed.

paulinaporizkova wrote:

DG wrote:It's not that awesome to vandalize it.

hahaha I can almost hear you saying this in person, shakin your head

I wrote that. Someone replaced "Your GPA" with "your balls," so I though it was appropriate.

dusk2k2 wrote:I'm not so good with math, but this seems to not work in determining class rank at my school. For example, my school gives A+ and all first year classes are curved between 3.0 and 3.3. Yet, using historical numbers, when I put an example gpa of 4.0, it gives a class rank of approximately top 6 percent. This simply cannot be. Heck, when I put the highest gpa in the class, it gives a percentile of top 3 percent. Can anyone explain how this math works?

It's not perfect, hence the title "estimator." It assumes normality and goes off of there. That said, unless your school curve is very far from the normal distribution, it should be fairly close. The probable answer here is that either the data point you are using to estimate is unrepresentative, or your curve is very far from the normal distribution.

nymario wrote:Here's a question: can you list the expected GPAs for the top 20 ranked students. Obviously this will be subject to a good amount of variance, but I'm curious as to how the top 5% is likely to shake out.

Assuming 400 students, Median is 3.177 (mean 3.2 (or slightly less) -- think 3.19-3.195)
10%=3.622
15%=3.511
20%=3.477
25%=3.410

Yeah, just fiddle with the numbers until you get top 5%, top 4.8% (or whatever's appropriate for your class size), etc. As you noted, this is obviously subject to a great deal of variance - this sort of prediction becomes more and more unstable at the tails (the top 1% is an outlier, after all).

edit...

Will this not be as accurate if you are using a data point that is farther from where your GPA is? For example, my school only gives quartiles, so the only cut off point I can know is the top 25% cutoff. Would it be more accurate if I knew what the top 10% cutoff was?

dusk2k2 wrote:Will this not be as accurate if you are using a data point that is farther from where your GPA is? For example, my school only gives quartiles, so the only cut off point I can know is the top 25% cutoff. Would it be more accurate if I knew what the top 10% cutoff was?

If you're closer to the 10% than the 25%, yeah. But you use what you've got. I'd like to stress that this is an estimator, not a crystal ball. The difference in precision between using the 25% and 10% estimator should not be considered concrete enough to change any plans you may have.

dailygrind wrote:

dusk2k2 wrote:Will this not be as accurate if you are using a data point that is farther from where your GPA is? For example, my school only gives quartiles, so the only cut off point I can know is the top 25% cutoff. Would it be more accurate if I knew what the top 10% cutoff was?

If you're closer to the 10% than the 25%, yeah. But you use what you've got. I'd like to stress that this is an estimator, not a crystal ball. The difference in precision between using the 25% and 10% estimator should not be considered concrete enough to change any plans you may have.

Just got informed of exact rank today, and yeah, the estimator is more accurate if you use the % that you are probably closer to.

hah didn't realize multiple people could edit it at the same time. the median number kept getting changed to 3.25 while i was inputting the other fields

traehekat wrote:

dailygrind wrote:

dusk2k2 wrote:Will this not be as accurate if you are using a data point that is farther from where your GPA is? For example, my school only gives quartiles, so the only cut off point I can know is the top 25% cutoff. Would it be more accurate if I knew what the top 10% cutoff was?

If you're closer to the 10% than the 25%, yeah. But you use what you've got. I'd like to stress that this is an estimator, not a crystal ball. The difference in precision between using the 25% and 10% estimator should not be considered concrete enough to change any plans you may have.

Just got informed of exact rank today, and yeah, the estimator is more accurate if you use the % that you are probably closer to.

how off was your actual ranka the estimator used here?

I would just point out that curves are highly unpredictable. I was talking to a guy who graduated from Georgetown a couple years ago, back when grades there were less inflated. He said that his 3.9, at the time, was just in the top 4%. Given inflation, I expected my 3.88 to be top 6%. (Top 10% is 3.71, so normally the 95th percentile would hit at about 3.85, maybe a bit lower, but if his 3.9 was just top 4%, I just assumed that the distribution at Georgetown must have an oddly high tail.) But in fact, it's top 1%, which makes no sense at all, as it means that 3.71-3.88 = top 10% to top 2%, while 3.88-4.0 = top 1%. (And for all I know, maybe a 3.86 is top 1% too; I have no reason to assume that I'm at the very bottom of the top 1%.) Now, it strikes me that both the current distribution, and the alumnus's distribution, are really weird. Normally you'd think our GPAs would be top 2%, top 3%, but not as low as top 4% and not as high as top 1%. I was so shocked when I got my rank that I asked the registrar if it was a typo. The moral of the story is, the estimator's probably way off and in any event, it doesn't matter because the only information that matters is the information you're allowed to report.

dailygrind wrote:

Hadlendale wrote:Anyone know a good method for obtaining an example data point? Or is this just something that is available via google for certain schools?

Word of mouth for what determines honors, I would imagine. My school has a law review grade-on for the top 25 students in our class of ~370, and typically number 25 falls somewhere in the area of 3.7 (this is the common lore of the school, anyway), so we use 3.7 and 25/370 = top 6.8% as our data point.

You can check the NALP directory of law schools. They list grade cutoffs for different designations 10%, 25: etc.. and the cut-offs for honors.

I started playing around with it to mentally masturbate to what grades I need for what rank, so I altered the estimator to an all-in-one gpa calculator + rank estimator.

Any other 0Ls can feel free to play with it at their own risk until someone messes it up: https://docs.google.com/spreadsheet/ccc ... UxHSllhNlE

-Feel free to change the class names, those don't really matter
-If you leave the grade field blank then it doesn't get factored into the GPA (so you can put nothing and it won't count as an F if you choose to not count 2nd semester courses)
-You can put letter grades (A, A+, A-, B+, etc) or point grades (4.0, 3.7, 3.3, etc) into the grade box, they both get treated the same (if you put a letter it gets translated as A+ = 4.3, A- = 3.7, B+ = 3.3, etc)

I only have my first semester grades. Does it make sense to even bother using this to calculate my rank since it's only done after second semester? I don't want to get ahead of myself or anything.

ilovesf wrote:I only have my first semester grades. Does it make sense to even bother using this to calculate my rank since it's only done after second semester? I don't want to get ahead of myself or anything.

Yeah, can't hurt. Especially if you're considering an ED transfer (just uses first semester grades).

So it won't let me change the values, nor does right clicking do anything.

niederbomb wrote:So it won't let me change the values, nor does right clicking do anything.

Should be good to go now. The editable class rank calc is the first page. If you click the tab at the bottom for the locked (protected copy) you can duplicate it. I've also integrated the combined GPA/Rank Calc into the spreadsheet.

.

Anybody know how to calculate class rank at a school like Berkeley? Grading scale within each subject is like this:

Top 10%: HH
11-40%= H
bottom 60% = P

Hypothetically ( ) if I got 3 H my first semester (out of 3 classes), does anybody know of a way to get some kind of rough estimate on what my class rank (by percentile) might be?

Thanks so much

ProsBuff wrote:Anybody know how to calculate class rank at a school like Berkeley? Grading scale within each subject is like this:

Top 10%: HH
11-40%= H
bottom 60% = P

Hypothetically ( ) if I got 3 H my first semester (out of 3 classes), does anybody know of a way to get some kind of rough estimate on what my class rank (by percentile) might be?

Thanks so much

I don't think it's possible without more info. If I had to guess, probably somewhere between top 8% and top 22%.

.

Jessuf wrote:

Jessuf wrote:So I found some GPAs for top 10% and top 25% from 2010. However, the curve at my school was 2.7-2.9 then. It is now 2.9-3.1. Is there any way to adjust this data to fit my GPA? I am a statistics dummy.

Ok, oops. The data I found was actually for the right median.

So why does the calculator keep telling me I'm ranked #1?! That's not possible with my GPA.

It's possible that it gets a little distorted at the extremes, so it probably won't be able to tell you your exact rank. But you can probably at least safely assume you're top 1-2%

.

Fuck. Why didn't I see this before? It requires so many assumptions, but it is a pretty fair estimate. Well done NYC.

I wish I knew any ranks at my school. Best I can do is guess that rank 1 is a 4.2.


3.76
3.24
4.2
1
10.38157256