Prepare for the LSAT or discuss it with others in this forum.
mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

I'm relatively new to the admissions game. it appears to me that most people here say that you have a chance at x school ONLY if you score higher than median. why's that? i mean 50% people get in to that school with a lower than median LSAT score. my guess is that as long as you have the 25 percentile, then you have a chance. am i wrong (a sincere question)?

i was bored and did some simple calculation. suppose there are 100,000 people taking LSAT each year (probably fewer?), since 170 is 99 percentile, then only 1000 people would score 170 or higher. then suppose there are another 1000 people scored 170+ in previous years applying in this cycle, then there are only 2000 people with 170+ in total applying this year. let's say the average number of seats for T14 is 300, then there will be 5200 seats opening up each year. so most people get into T14 with a score lower than 170? But this is just my guess, I have no idea if any of these numbers is a good estimate.

revised based on Steve's new numbers:
sure. let's revise the number then. 1800 score 170+ this year, 1500 score 170+ from previous year, still only 3300 students score than 170+ applying.

based on the following data, 4550 seats in T14, and 6450 seats in T20. So I guess 25% people get in T14 with 170- and about 50% for T20?

also consider the following:

1 some people don't care about T14. they want to attend schools close to home.
2 some people take scholarships from a lower ranking school.
3 some have low GPA or other negative elements that will overall affect their application strenghth.

1
Harvard University
173.0
3.855
550

2
Yale University
173.0
3.890
200

3
Columbia University
172.5
3.700
400

4
New York University
171.0
3.705
450

5
University of Chicago
171.0
3.680
200

6
Stanford University
170.0
3.850
200

7
University of Virginia
168.5
3.725
350

8
Northwestern University
169.0
3.605
250

9
University of Pennsylvania
168.5
3.670
250

10
University of Michigan, Ann Arbor
168.0
3.730
350

11
Georgetown University
168.0
3.610
600

12
University of California, Berkeley
167.0
3.810
250

13
Duke University
167.5
3.700
200

14
University of California, Los Angeles
166.5
3.680
300

15
Vanderbilt University
166.5
3.690
200

16
Cornell University
167.0
3.540
200

17
University of Texas, Austin
166.0
3.620
450

18
University of Southern California
166.0
3.605
200

19
George Washington University
165.0
3.555
600

20
Emory University
165.0
3.565
250
Last edited by mz253 on Fri Jun 04, 2010 2:02 pm, edited 1 time in total.

LSAT Blog

Posts: 1257
Joined: Mon Dec 07, 2009 9:24 pm

Two resources that can help with your calculations:

http://www.lawschoolpredictor.com

Enjoy.

-Steve

09042014

Posts: 18204
Joined: Wed Oct 14, 2009 10:47 pm

mz253 wrote:I'm relatively new to the admissions game. it appears to me that most people here say that you have a chance at x school ONLY if you score higher than median. why's that? i mean 50% people get in to that school with a lower than median LSAT score. my guess is that as long as you have the 25 percentile, then you have a chance. am i wrong (a sincere question)?

i was bored and did some simple calculation. suppose there are 100,000 people taking LSAT each year (probably fewer?), since 170 is 99 percentile, then only 1000 people would score 170 or higher. then suppose there are another 1000 people scored 170+ in previous years applying in this cycle, then there are only 2000 people with 170+ in total applying this year. let's say the average number of seats for T14 is 300, then there will be 5200 seats opening up each year. so most people get into T14 with a score lower than 170? But this is just my guess, I have no idea if any of these numbers is a good estimate.

You absolutely need an above median LSAT or an above median GPA. 170 is 97.5% not 99%.

As to why they require a near median LSAT, its because they don't want dumber students.

Flanker1067

Posts: 658
Joined: Tue Dec 29, 2009 12:47 pm

So much fail in this post. Aside from you being no where close on your calculations of 170+ scorers, US news considers medians in their rankings. Since US news cares, schools care, so they median protect/try to up their medians.

Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

sure. let's revise the number then. 1800 score 170+ this year, 1500 score 170+ from previous year, still only 3300 students score than 170+ applying.

based on the following data, 4550 seats in T14, and 6450 seats in T20. So I guess 25% people get in T14 with 170- and about 50% for T20?

also consider the following:

1 some people don't care about T14. they want to attend schools close to home.
2 some people take scholarships from a lower ranking school.
3 some have low GPA or other negative elements that will overall affect their application strenghth.

1
Harvard University
173.0
3.855
550

2
Yale University
173.0
3.890
200

3
Columbia University
172.5
3.700
400

4
New York University
171.0
3.705
450

5
University of Chicago
171.0
3.680
200

6
Stanford University
170.0
3.850
200

7
University of Virginia
168.5
3.725
350

8
Northwestern University
169.0
3.605
250

9
University of Pennsylvania
168.5
3.670
250

10
University of Michigan, Ann Arbor
168.0
3.730
350

11
Georgetown University
168.0
3.610
600

12
University of California, Berkeley
167.0
3.810
250

13
Duke University
167.5
3.700
200

14
University of California, Los Angeles
166.5
3.680
300

15
Vanderbilt University
166.5
3.690
200

16
Cornell University
167.0
3.540
200

17
University of Texas, Austin
166.0
3.620
450

18
University of Southern California
166.0
3.605
200

19
George Washington University
165.0
3.555
600

20
Emory University
165.0
3.565
250

lostjake

Posts: 317
Joined: Sat Feb 06, 2010 7:07 pm

Flanker1067 wrote:So much fail in this post. Aside from you being no where close on your calculations of 170+ scorers, US news considers medians in their rankings. Since US news cares, schools care, so they median protect/try to up their medians.

Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

fail.

mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

sorry that i don't know they rank based on median score.

so as for the median score, if 10 accepted students are like this:

161 162 163 164 170 170 170 172 173 175

then the median is 170 right (is median the number in the middle, or if there are two numbers then median is the average)? so if a school only cares about median, then if you are below median, then they don't care about how low you go?

Flanker1067 wrote:So much fail in this post. Aside from you being no where close on your calculations of 170+ scorers, US news considers medians in their rankings. Since US news cares, schools care, so they median protect/try to up their medians.

Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

also, since they rank by median, so the most popular score for top schools would be 173 174 175 right? and if you hit 178+, it will not make a significant difference for their ranking because 178 is almost impossible to be their median score.

or i'm just too dumb to make that assumption up?

Flanker1067

Posts: 658
Joined: Tue Dec 29, 2009 12:47 pm

Still don't know where those numbers are coming from. Did you read the first response? 170,000 (roughly) took the lsat this last cycle.

Edit: there is the big question though of how many unique takers there are. This is anyones guess.
Last edited by Flanker1067 on Fri Jun 04, 2010 2:02 pm, edited 1 time in total.

mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

yea.

i just revised my assumption... sorry.

Flanker1067 wrote:Still don't know where those numbers are coming from. Did you read the first response? 170,000 (roughly) took the lsat this last cycle.

Flanker1067

Posts: 658
Joined: Tue Dec 29, 2009 12:47 pm

lostjake wrote:
Flanker1067 wrote:So much fail in this post. Aside from you being no where close on your calculations of 170+ scorers, US news considers medians in their rankings. Since US news cares, schools care, so they median protect/try to up their medians.

Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

fail.

What part of that is wrong?

mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

woops, i forgot to count retakers. how many percent of 170+ succeeded in first try?

Flanker1067

Posts: 658
Joined: Tue Dec 29, 2009 12:47 pm

mz253 wrote:yea.

i just revised my assumption... sorry.

Flanker1067 wrote:Still don't know where those numbers are coming from. Did you read the first response? 170,000 (roughly) took the lsat this last cycle.

Huh? you need to assume 170K (you can adjust for some sort of estimate of how many are retakers/don't apply) at 170 being the 97.5 percentile.

quishiclocus

Posts: 45
Joined: Fri May 07, 2010 5:55 pm

I think you'll find that most people are not in the exact same part of the distribution for both GPA and LSAT. If you have a 25th percentile LSAT, but a 75th percentile plus GPA, that's going to offset to a certain degree, although not necessarily to the same degree as being roughly median in both. Just like if you have a high LSAT score and a below-median GPA that may offset to a degree. But those people with below-median LSAT are not being admitted because of their LSAT. They're being admitted because the rest of their application is strong enough to carry a 'meh' LSAT. This is not, given that the LSAT is usually the last major factor you can control, the position you want to be in.

LSAT Blog

Posts: 1257
Joined: Mon Dec 07, 2009 9:24 pm

mz253 wrote:woops, i forgot to count retakers.

You may also want to make some sort of allowance for those who take the LSAT and then end up not applying to law school.

Edit: Flanker beat me to it.

mz253 wrote:how many percent of 170+ succeeded in first try?

This data on retaker stats is buried somewhere on LSAC's website.

Edit: Found it: http://www.lsac.org/pdfs/RepeaterData.pdf
Last edited by LSAT Blog on Fri Jun 04, 2010 2:09 pm, edited 1 time in total.

mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

i agree. since my GPA is in every school's 75 percentile, so i think if i hit median, then i'm in. but i still think if i hit 25 percentile, i have some chance.

maybe i'm too stressed out about Monday. i keep telling myself i just need to score high 160s and i'm happy to go to a T20 or even a T30 school.

quishiclocus wrote:I think you'll find that most people are not in the exact same part of the distribution for both GPA and LSAT. If you have a 25th percentile LSAT, but a 75th percentile plus GPA, that's going to offset to a certain degree, although not necessarily to the same degree as being roughly median in both. Just like if you have a high LSAT score and a below-median GPA that may offset to a degree. But those people with below-median LSAT are not being admitted because of their LSAT. They're being admitted because the rest of their application is strong enough to carry a 'meh' LSAT. This is not, given that the LSAT is usually the last major factor you can control, the position you want to be in.

mz253

Posts: 319
Joined: Thu Mar 25, 2010 11:18 pm

well, i'm trying to make the scenario as miserable as possible. but if 170 is only 97.5%, then there should be 180000*2.5%= 4500 score higher than 170. but anyway you are right, lots of people prob won't be up for a top school or do not apply, and 170+ from previous years probably won't be as high as 1500?

LSAT Blog wrote:
mz253 wrote:woops, i forgot to count retakers.

You may also want to make some sort of allowance for those who take the LSAT and then end up not applying to law school.

Edit: Flanker beat me to it.

mz253 wrote:how many percent of 170+ succeeded in first try?

This data on retaker stats is buried somewhere on LSAC's website.

Flanker1067

Posts: 658
Joined: Tue Dec 29, 2009 12:47 pm

mz253 wrote:i agree. since my GPA is in every school's 75 percentile, so i think if i hit median, then i'm in. but i still think if i hit 25 percentile, i have some chance.

maybe i'm too stressed out about Monday. i keep telling myself i just need to score high 160s and i'm happy to go to a T20 or even a T30 school.

quishiclocus wrote:I think you'll find that most people are not in the exact same part of the distribution for both GPA and LSAT. If you have a 25th percentile LSAT, but a 75th percentile plus GPA, that's going to offset to a certain degree, although not necessarily to the same degree as being roughly median in both. Just like if you have a high LSAT score and a below-median GPA that may offset to a degree. But those people with below-median LSAT are not being admitted because of their LSAT. They're being admitted because the rest of their application is strong enough to carry a 'meh' LSAT. This is not, given that the LSAT is usually the last major factor you can control, the position you want to be in.

This is basically true. Congrats on having a good GPA, it may even get you in somewhere that your LSAT is below median. Anyway, good luck, just kill it and then you don't have to worry about medians.

lostjake

Posts: 317
Joined: Sat Feb 06, 2010 7:07 pm

Flanker1067 wrote:
lostjake wrote:
Flanker1067 wrote:So much fail in this post. Aside from you being no where close on your calculations of 170+ scorers, US news considers medians in their rankings. Since US news cares, schools care, so they median protect/try to up their medians.

Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

fail.

What part of that is wrong?

50% of the class will have below median LSAT scores
50% of the class will have below median GPA

Maybe you're thinking of the mean

d34d9823

Posts: 1879
Joined: Wed Apr 14, 2010 2:52 pm

Flanker1067 wrote:Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

Uh...yes they do, that's the definition.

balzern

Posts: 347
Joined: Tue May 25, 2010 2:27 pm

d34dluk3 wrote:
Flanker1067 wrote:Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

Uh...yes they do, that's the definition.

+1 lol

thechee

Posts: 174
Joined: Tue Oct 06, 2009 8:42 am

I responded to another version of this post here: viewtopic.php?f=7&t=119409

thechee wrote:Last year over 170,000 LSATs were administered. If we assume that 30% of these are repeaters, then that gives us c120,000 unique LSAT takers. Source: LSAC data on lsac.org

170 is the 97.5 percentile, so that gives us at least 3000 test takers scoring a 170 or above. Test takers from previous years likely don't change the numbers that much, since many taking it this year will also choose to use their scores further down the road (although ITE it's likely that fewer people are waiting than in past years, given crappy post-college job prospects).

So, there are around 4500 1st year seats in the T14, give or take. A decent proportion of those with 170+ scores will likely be shut out because of low GPAs, although some of the splitter-friendly schools (NU, UVA) might take a handful of them. Still, I think it's safe to say that at least 2000 (but probably more) T14 seats are going to those with 170+ scores.

That leaves 2500 seats.

Now, at 169 (96.8 percentile), you have about 1000 test takers, and another 850 at 168 (96.1). I won't speculate about what % of these individuals crack the T14, but I don't think 50% is too liberal. For argument's sake, let's say 1000 of the 1850 at 168-169 make it into the T14. Now were are at 1500 seats left.*

So, if at least 2/3 of the T14 can be filled by those with 168+ scores, it seems obvious that the score alone is a pretty powerful determinant of your admissions chances. For each point below 168, you have more and more people with the same score, making it even harder to distinguish between applicants. For example, there are nearly as many people with a 163 as there are with a 170-180.

I'm not saying that it's impossible to overcome a sub-25th% LSAT score for a given school (a good buddy of mine got into a T10 with a 160/non-URM), but the odds are definitely not in your favor. Grades can help, but that still doesn't change the fact that the supply of high GPAs is much great that the supply of high LSAT scores.

*Then of course there are URMs, who tend to get a significant boost, despite more often than not having LSAT scores lower than those of other non-URMs admitted to the same school. Again, not sure what the percentage is here, but I think we can assume that it's significant enough to significantly reduce the number of available seats.

rx3r

Posts: 225
Joined: Mon Jun 15, 2009 9:35 pm

balzern wrote:
d34dluk3 wrote:
Flanker1067 wrote:Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

Uh...yes they do, that's the definition.

+1 lol

False, see example: 11 LSAT scores 170, 171, 171, 171, 171, 171, 171, 172, 172, 172, 180

The median here is 171. However, 50% did not score UNDER 171.

09042014

Posts: 18204
Joined: Wed Oct 14, 2009 10:47 pm

lostjake wrote:
Flanker1067 wrote:
lostjake wrote:
Flanker1067 wrote:So much fail in this post. Aside from you being no where close on your calculations of 170+ scorers, US news considers medians in their rankings. Since US news cares, schools care, so they median protect/try to up their medians.

Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

fail.

What part of that is wrong?

50% of the class will have below median LSAT scores
50% of the class will have below median GPA

Maybe you're thinking of the mean

50+% of the class will be AT median or below.

balzern

Posts: 347
Joined: Tue May 25, 2010 2:27 pm

rx3r wrote:
balzern wrote:
d34dluk3 wrote:
Flanker1067 wrote:Edit: Also, please learn what a median is. 50% of a class does NOT have below median scores.

Uh...yes they do, that's the definition.

+1 lol

False, see example: 11 LSAT scores 170, 171, 171, 171, 171, 171, 171, 172, 172, 172, 180

The median here is 171. However, 50% did not score UNDER 171.

50 % scored AT OR BELLOW a 171.