Class sizes and medians (c/o 2015) Forum

[Betharl]

Wow @ CLS. They maintained their LSAT median (75th fell), but median GPA fell. And that's with a 9.4% decrease in class size.

ETA: Considering NYU did not reduce class size, I'm calling bullshit if they post a 172 median.

I thought people had already said NYU's median fell to 171? Or were they just rumors (or am I crazy... quite possible)?

Pretty sure it was rumors but it would seem to be impossible that the median stayed at 172.

Also, is there evidence that NYU was more liberal with scholarship money and maybe poached people that would have otherwise gone to CLS?

guys am i retarded math-wise? how can the median NOT be 171 if it's the middle point between the 75th and 25th percentiles (i.e. 173...172...171...170...169)???

It isn't the middle point between the 75th and 25th percentiles, it could equal the 25th or 75th or be anywhere in between. Typically, the way the numbers are distributed with law schools, the median is closer to the 75th.

EDIT: See Kevin vvv

[KevinP]

Wow @ CLS. They maintained their LSAT median (75th fell), but median GPA fell. And that's with a 9.4% decrease in class size.

ETA: Considering NYU did not reduce class size, I'm calling bullshit if they post a 172 median.

I thought people had already said NYU's median fell to 171? Or were they just rumors (or am I crazy... quite possible)?

That's what people said last year... so i'm a bit skeptical.

guys am i retarded math-wise? how can the median NOT be 171 if it's the middle point between the 75th and 25th percentiles (i.e. 173...172...171...170...169)???

Because LSAT scores can repeat and have gaps.

Here... 100 students:
26 have 173
48 have 171
26 have 169


75th = 173
median = 171
25th = 169

[danquayle]

Wow @ CLS. They maintained their LSAT median (75th fell), but median GPA fell. And that's with a 9.4% decrease in class size.

ETA: Considering NYU did not reduce class size, I'm calling bullshit if they post a 172 median.

I thought people had already said NYU's median fell to 171? Or were they just rumors (or am I crazy... quite possible)?

Pretty sure it was rumors but it would seem to be impossible that the median stayed at 172.

Also, is there evidence that NYU was more liberal with scholarship money and maybe poached people that would have otherwise gone to CLS?

guys am i retarded math-wise? how can the median NOT be 171 if it's the middle point between the 75th and 25th percentiles (i.e. 173...172...171...170...169)???

All median means is that half the scores are higher, and half the scores are lower. It doesn't matter how much lower or higher each score is, simply that they either are or they are not higher. So, say you have 13 scores. Whatever your 7th lowest score will be your median. So if you want your median to be 172, you just need to make sure you have at least seven scores greater than or equal to 172. The rest of the scores could be 6 180s and 6 150s... it won't matter. (But if 2 180s drop, your median suddenly becomes 150).

177
176
173
173
173
172

Median = 172 (6 scores higher, 6 scores lower)

171
169
169
169
163
162

In this distribution, the 75th = 173, 25th = 169, with a 172 median.

[99.9luft]

Kev, Dan - thank you! That makes complete sense, i should have considered the distribution of those scores throughout the demographics!

[Betharl]

I'm interested to know what the decrease in apps will be this cycle. There were 6% less LSATs administered in June, which is less than the decrease last cycle, but we should expect to see a greater reduction in apps due to the "lag effect" than what was experienced last cycle.

By lag effect, I mean people applying this cycle with LSAT scores from last cycle. For example, I took three LSATS, Dec'11, Feb'12, and June'12. I will be applying this cycle, but 2/3 of my LSATS were taken last cycle. There should be less people like me out there this cycle compared to last cycle since a lot more people took the lsat in the year before the last cycle than this one. But how much impact will this have?

[thelawyler]

I'm interested to know what the decrease in apps will be this cycle. There were 6% less LSATs administered in June, which is less than the decrease last cycle, but we should expect to see a great reduction in apps due to the "lag effect" than what was experienced last cycle.

By lag effect, I mean people applying this cycle with LSAT scores from last cycle. For example, I took three LSATS, Dec'11, Feb'12, and June'12. I will be applying this cycle, but 2/3 of my LSATS were taken last cycle. There should be less people like me out there this cycle compared to last cycle since a lot more people took the lsat in the year before the last cycle than this one. But how much impact will this have?

Well, people who take the LSAT this year will also wait another year like you did, so I'd say its a wash.

[danquayle]

To make it easier to see that quartiles work the same way, heres how the hypothetical 173, 172, 169 would work if you expanded to 15 scores:

177
176
173

173 = 75th (1/4 higher or equal, 3/4 lower or equal)

173
173
172

Median = 172 (1/2 scores higher or equal, 1/2 scores lower or equal)

171
169
169

169 = 25th (3/4 higher or equal, 1/4 lower or equal)

164
155
150


75th is the 4th lowest score, the median the 8th lowest score (7 higher, 7 lower) and the 25th the 12th lowest score.

[Betharl]

I'm interested to know what the decrease in apps will be this cycle. There were 6% less LSATs administered in June, which is less than the decrease last cycle, but we should expect to see a great reduction in apps due to the "lag effect" than what was experienced last cycle.

By lag effect, I mean people applying this cycle with LSAT scores from last cycle. For example, I took three LSATS, Dec'11, Feb'12, and June'12. I will be applying this cycle, but 2/3 of my LSATS were taken last cycle. There should be less people like me out there this cycle compared to last cycle since a lot more people took the lsat in the year before the last cycle than this one. But how much impact will this have?

Well, people who take the LSAT this year will also wait another year like you did, so I'd say its a wash.

Yeah, but as LSATs administered goes down, there should also be less reappliers. So, last cycle, there were probably more reappliers than there will be this cycle. This should result in a reduction of apps, but I don't know to what degree. We should have less takers (if the June trend of -6% continues) and less reappliers than last cycle. Overall, maybe the drop in apps this cycle will approximately equal the drop in apps from last cycle, even though the drop in LSATs administered isn't expected to be as high (with the "lag effect" making up the difference).

[KevinP]

I'm interested to know what the decrease in apps will be this cycle. There were 6% less LSATs administered in June, which is less than the decrease last cycle, but we should expect to see a greater reduction in apps due to the "lag effect" than what was experienced last cycle.

By lag effect, I mean people applying this cycle with LSAT scores from last cycle. For example, I took three LSATS, Dec'11, Feb'12, and June'12. I will be applying this cycle, but 2/3 of my LSATS were taken last cycle. There should be less people like me out there this cycle compared to last cycle since a lot more people took the lsat in the year before the last cycle than this one. But how much impact will this have?

It's going to be hard to predict. Thing is, schools did not expect such drastic declines and therefore many coped with the decline in unsustainable ways (e.g. decrease in class sizes or massive scholly money). Some schools will def. be able to keep that up this cycle, but many also won't.

[Tom Joad]

What's with CLS and NYU bombing out on median GPA?

[ajax]

I'm interested to know what the decrease in apps will be this cycle. There were 6% less LSATs administered in June, which is less than the decrease last cycle, but we should expect to see a greater reduction in apps due to the "lag effect" than what was experienced last cycle.

By lag effect, I mean people applying this cycle with LSAT scores from last cycle. For example, I took three LSATS, Dec'11, Feb'12, and June'12. I will be applying this cycle, but 2/3 of my LSATS were taken last cycle. There should be less people like me out there this cycle compared to last cycle since a lot more people took the lsat in the year before the last cycle than this one. But how much impact will this have?

It's going to be hard to predict. Thing is, schools did not expect such drastic declines and therefore many coped with the decline in unsustainable ways (e.g. decrease in class sizes or massive scholly money). Some schools will def. be able to keep that up this cycle, but many also won't.

The main issue is we have to take these schools at their word. The ones that are reporting no drop in median while keeping class size the same seem extremely suspect. What is to prevent them from pulling an Illinois/Villanova?

[VUSisterRayVU]

I'm interested to know what the decrease in apps will be this cycle. There were 6% less LSATs administered in June, which is less than the decrease last cycle, but we should expect to see a greater reduction in apps due to the "lag effect" than what was experienced last cycle.

By lag effect, I mean people applying this cycle with LSAT scores from last cycle. For example, I took three LSATS, Dec'11, Feb'12, and June'12. I will be applying this cycle, but 2/3 of my LSATS were taken last cycle. There should be less people like me out there this cycle compared to last cycle since a lot more people took the lsat in the year before the last cycle than this one. But how much impact will this have?

It's going to be hard to predict. Thing is, schools did not expect such drastic declines and therefore many coped with the decline in unsustainable ways (e.g. decrease in class sizes or massive scholly money). Some schools will def. be able to keep that up this cycle, but many also won't.

The main issue is we have to take these schools at their word. The ones that are reporting no drop in median while keeping class size the same seem extremely suspect. What is to prevent them from pulling an Illinois/Villanova?

Reputation. Look what happened to those schools. I HIGHLY doubt any T14 is lying on stats. Obviously gaming the system occurs, but nothing as scandalous as Illinois.

@Kevin, what do you suspect the T14 will do this time? Do you expect them to just decrease class sizes to boost medians thereby making this cycle less juicy?

[justonemoregame]

The t14s that aren't splitter-friendly could become a little more-so. If we assume the 6% decline for the next few LSAT administrations, can we estimate how many 169/170+ LSATs are out there for the upcoming cycle?

If we have 120,000 tests, 3.5% of that is like 4,200. That seems high

[BigZuck]

Duke is already outright accepting splitters and reverse splitters and people under both medians. Seems unprecedented from what I can tell. Epicness is coming...

[TheThriller]

Duke is already outright accepting splitters and reverse splitters and people under both medians. Seems unprecedented from what I can tell. Epicness is coming...

What was Duke's median LSAT last year?

[iamrobk]

I'm interested to know what the decrease in apps will be this cycle. There were 6% less LSATs administered in June, which is less than the decrease last cycle, but we should expect to see a greater reduction in apps due to the "lag effect" than what was experienced last cycle.

By lag effect, I mean people applying this cycle with LSAT scores from last cycle. For example, I took three LSATS, Dec'11, Feb'12, and June'12. I will be applying this cycle, but 2/3 of my LSATS were taken last cycle. There should be less people like me out there this cycle compared to last cycle since a lot more people took the lsat in the year before the last cycle than this one. But how much impact will this have?

It's going to be hard to predict. Thing is, schools did not expect such drastic declines and therefore many coped with the decline in unsustainable ways (e.g. decrease in class sizes or massive scholly money). Some schools will def. be able to keep that up this cycle, but many also won't.

The main issue is we have to take these schools at their word. The ones that are reporting no drop in median while keeping class size the same seem extremely suspect. What is to prevent them from pulling an Illinois/Villanova?

Reputation. Look what happened to those schools. I HIGHLY doubt any T14 is lying on stats. Obviously gaming the system occurs, but nothing as scandalous as Illinois.

@Kevin, what do you suspect the T14 will do this time? Do you expect them to just decrease class sizes to boost medians thereby making this cycle less juicy?

Just one example, but I think Penn's class went from 266 to 244.

[BigZuck]

Duke is already outright accepting splitters and reverse splitters and people under both medians. Seems unprecedented from what I can tell. Epicness is coming...

What was Duke's median LSAT last year?

3.75/170. Not sure what it is now though, don't think it has been released.

[dowu]

Duke is already outright accepting splitters and reverse splitters and people under both medians. Seems unprecedented from what I can tell. Epicness is coming...

How do you know that Duke has been accepting splitters and people below both medians, etc...?

[Funkycrime]

Duke is already outright accepting splitters and reverse splitters and people under both medians. Seems unprecedented from what I can tell. Epicness is coming...

How do you know that Duke has been accepting splitters and people below both medians, etc...?

probably referring to the application thread.

[BigZuck]

I don't really care to share my numbers but I was accepted and I am around one median and well below another.

[dowu]

I don't really care to share my numbers but I was accepted and I am around one median and well below another.

Around meaning below or above? Did you get a scholly?

[thelawyler]

I don't really care to share my numbers but I was accepted and I am around one median and well below another.

Around meaning below or above? Did you get a scholly?

It means GPA is close enough to be a wash, and LSAT was much lower.

[2014]

UChi sections are out and the early look suggests we will have 186 students which is down 5? from last year. Should know medians next week.

[20130312]

Rutgers Camden hasn't released any of these numbers and hasn't put out a timetable to release them (shockingly enough).

[dowu]

I don't really care to share my numbers but I was accepted and I am around one median and well below another.

Around meaning below or above? Did you get a scholly?

It means GPA is close enough to be a wash, and LSAT was much lower.

Hmm, I guess I should have been specific. I'm wondering whether schools are throwing money around like it's going out of style, for applicants like Bigzuck.