Law School Predictors and Calculators Forum

[DoubleChecks]

So I know about LSP and LSN, etc. but today I was having a discussion with my friend when he mentioned how many schools submit ALL their incoming student numbers (whole shebang, not just 25/75) to LSAC or something. In turn, some sites could 'buy' or get their hands on these complete sets of numbers and create their own predictors. Was he mistaken in some way? I mean, if these exist, why bother with LSP? Wouldn't these be, for all intents and purposes, 100% accurate for x school of x yr when it came to numbers?

[DearCan]

If I'm understand your question correctly, LSPs and calculators just make it convenient for an applicant to plug in their specific numbers and see the potential outcomes. It's more convenient than sifting through a bunch of previous applicants' numbers.

[CanadianWolf]

Don't know. Rhetorical question.Yes.

[DoubleChecks]

If I'm understand your question correctly, LSPs and calculators just make it convenient for an applicant to plug in their specific numbers and see the potential outcomes. It's more convenient than sifting through a bunch of previous applicants' numbers.

well yeah, but i mean the 'potential outcomes' are based on something, say past LSN data for the last 10 yrs

my friend is saying there is an ACCESSIBLE amount of data that is the ENTIRE incoming class stats for almost all schools (given to LSAC) - is this true? and if so, why wouldnt LSP use this info?

[sundance95]

Because its written by a TLSer who has better things to do then plug in #s for every newly matriculating law student in the United States? (I believe it is YCRevolution, plz correct me if I am wrong) And this is assuming your friend is correct regarding the availability of this information, which I believe is inaccurate.

[DearCan]

If I'm understand your question correctly, LSPs and calculators just make it convenient for an applicant to plug in their specific numbers and see the potential outcomes. It's more convenient than sifting through a bunch of previous applicants' numbers.

well yeah, but i mean the 'potential outcomes' are based on something, say past LSN data for the last 10 yrs

my friend is saying there is an ACCESSIBLE amount of data that is the ENTIRE incoming class stats for almost all schools (given to LSAC) - is this true? and if so, why wouldnt LSP use this info?

Ah, I see. Yeah, most of these predictions are based on a self-selected sample. I don't know what LSP uses, but the LSAC probability chart is useful, and probably the most reliable. Although, if I'm to understand correctly, not all schools report this data to LSAC so for those schools there's really not much to go on other than past cycles and info from LSN and such.

If I'm completely misunderstanding, forgive me.

[DoubleChecks]

If I'm understand your question correctly, LSPs and calculators just make it convenient for an applicant to plug in their specific numbers and see the potential outcomes. It's more convenient than sifting through a bunch of previous applicants' numbers.

well yeah, but i mean the 'potential outcomes' are based on something, say past LSN data for the last 10 yrs

my friend is saying there is an ACCESSIBLE amount of data that is the ENTIRE incoming class stats for almost all schools (given to LSAC) - is this true? and if so, why wouldnt LSP use this info?

Ah, I see. Yeah, most of these predictions are based on a self-selected sample. I don't know what LSP uses, but the LSAC probability chart is useful, and probably the most reliable. Although, if I'm to understand correctly, not all schools report this data to LSAC so for those schools there's really not much to go on other than past cycles and info from LSN and such.

If I'm completely misunderstanding, forgive me.

what is this probability chart you speak of?

[r6_philly]

I do not believe admission outcomes can be sold by LSAC or the law schools. But I am not a privacy lawyer, yet.

I deal with trafficking user/member data all day, and my instinct says they wouldn't have the authority to sell it or use it commercially.

[DoubleChecks]

I do not believe admission outcomes can be sold by LSAC or the law schools. But I am not a privacy lawyer, yet.

I deal with trafficking user/member data all day, and my instinct says they wouldn't have the authority to sell it or use it commercially.

ok lets just say create their own predictor

[r6_philly]

I do not believe admission outcomes can be sold by LSAC or the law schools. But I am not a privacy lawyer, yet.

I deal with trafficking user/member data all day, and my instinct says they wouldn't have the authority to sell it or use it commercially.

ok lets just say create their own predictor

What if they are wrong? They are the authority to control admissions, they can't offer a predictor. A third party could, but LSAC wouldn't be able to release the data to third party for commercial purposes.

Well actually they could, but they would get sued. (I have been dealing with something strikingly similar, and we decided to not sell the data)

[Knock]

I do not believe admission outcomes can be sold by LSAC or the law schools. But I am not a privacy lawyer, yet.

I deal with trafficking user/member data all day, and my instinct says they wouldn't have the authority to sell it or use it commercially.

ok lets just say create their own predictor

No, because (especially at the higher ranked schools), there may be a person with X numbers who gets in, and a person who doesn't. You wouldn't really get that much more accurate than we are now, especially with so many data points on LSN and multiple cycles under TLS' belt.

[r6_philly]

I would go one step further and say there is not a good business model for anyone who would buy this data if they could. You can't really control access to this and charge for access, because people can just try out all kinds of numbers and relay that info to friends. You can't control the consumption of the service to make sure only the paying customer is consuming it. So that's why only LSAC would be the logical party to offer such as service, but they 1) can't and 2) gains nothing since they don't charge.

[DoubleChecks]

I do not believe admission outcomes can be sold by LSAC or the law schools. But I am not a privacy lawyer, yet.

I deal with trafficking user/member data all day, and my instinct says they wouldn't have the authority to sell it or use it commercially.

ok lets just say create their own predictor

No, because (especially at the higher ranked schools), there may be a person with X numbers who gets in, and a person who doesn't. You wouldn't really get that much more accurate than we are now, especially with so many data points on LSN and multiple cycles under TLS' belt.

sry i sort of don't follow here

the data would be a lot more accurate because it would be the ACTUAL data...still just a predictor of course, but instead of a biased sample size (LSN), it'd be the population essentially

[DoubleChecks]

I would go one step further and say there is not a good business model for anyone who would buy this data if they could. You can't really control access to this and charge for access, because people can just try out all kinds of numbers and relay that info to friends. You can't control the consumption of the service to make sure only the paying customer is consuming it. So that's why only LSAC would be the logical party to offer such as service, but they 1) can't and 2) gains nothing since they don't charge.

makes sense

[Knock]

I do not believe admission outcomes can be sold by LSAC or the law schools. But I am not a privacy lawyer, yet.

I deal with trafficking user/member data all day, and my instinct says they wouldn't have the authority to sell it or use it commercially.

ok lets just say create their own predictor

No, because (especially at the higher ranked schools), there may be a person with X numbers who gets in, and a person who doesn't. You wouldn't really get that much more accurate than we are now, especially with so many data points on LSN and multiple cycles under TLS' belt.

sry i sort of don't follow here

the data would be a lot more accurate because it would be the ACTUAL data...still just a predictor of course, but instead of a biased sample size (LSN), it'd be the population essentially

IE, we don't need every single last number to know that you have a good chance of admissions with median numbers, or you have an okay chance with XX combination of numbers, or you're in with GPA/LSAT above 75th%, etc. We pretty much have a good idea as of now, with the exception of URMs and splitters to some degree; but by their nature they will be unpredictable anyways...

Also, there is more to the admissions process than merely numbers as well, even if it doesn't seem so at times.

[2014]

I do not believe admission outcomes can be sold by LSAC or the law schools. But I am not a privacy lawyer, yet.

I deal with trafficking user/member data all day, and my instinct says they wouldn't have the authority to sell it or use it commercially.

ok lets just say create their own predictor

What if they are wrong? They are the authority to control admissions, they can't offer a predictor. A third party could, but LSAC wouldn't be able to release the data to third party for commercial purposes.

Well actually they could, but they would get sued. (I have been dealing with something strikingly similar, and we decided to not sell the data)

They would never be wrong.
They would set up the calculator that very clearly stated "Based on previous cycle, results in this cycle could vary" and instead of doing "Admit" "Strong Consider" etc, they could just do percentages. Even if someone has a 99% chance of admittance, if they don't get in all LSAC has to do is say they were the 1% and the results weren't guaranteed. Done deal.
I kind of wish they would release one too, but it would be an official acknowledgement that it is all numbers, and the law schools still want everyone to believe softs matter.

[Pleasye]

LSAC has something like LSP on their website...I forgot where I found it. But it's essentially the same thing: you put in a GPA and LSAT and they tell you your chances of getting in.

[DoubleChecks]

LSAC has something like LSP on their website...I forgot where I found it. But it's essentially the same thing: you put in a GPA and LSAT and they tell you your chances of getting in.

what #'s are they basing it off of? actual numbers the schools submit? cuz that would make LSP obsolete

im not actually debating about the importance of numbers...or whats the best calculator or how they work, etc.

my friend just said schools gave all incoming data to LSAC and that it was posted -- i had never heard of this and so i wanted to know if it were true or not, thats it

[Pleasye]

LSAC has something like LSP on their website...I forgot where I found it. But it's essentially the same thing: you put in a GPA and LSAT and they tell you your chances of getting in.

what #'s are they basing it off of? actual numbers the schools submit? cuz that would make LSP obsolete

im not actually debating about the importance of numbers...or whats the best calculator or how they work, etc.

my friend just said schools gave all incoming data to LSAC and that it was posted -- i had never heard of this and so i wanted to know if it were true or not, thats it

I have absolutely no idea what numbers it's based off of. If I track it down on the site again I'll see if it says...

[krizzly7]

LSAC has something like LSP on their website...I forgot where I found it. But it's essentially the same thing: you put in a GPA and LSAT and they tell you your chances of getting in.

what #'s are they basing it off of? actual numbers the schools submit? cuz that would make LSP obsolete

im not actually debating about the importance of numbers...or whats the best calculator or how they work, etc.

my friend just said schools gave all incoming data to LSAC and that it was posted -- i had never heard of this and so i wanted to know if it were true or not, thats it

I have absolutely no idea what numbers it's based off of. If I track it down on the site again I'll see if it says...

http://officialguide.lsac.org/release/O ... fault.aspx

[rinkrat19]

The LSAC predictor's disclaimer:

Results for the UGPA/LSAT search are computed using a logistic regression model employing Fall 2009 full-time application and admission data as reported by all ABA -approved law schools. The results presented represent a 95% probability that an applicant to a particular school for Fall 2009 would have had an actual chance of admission within the range indicated. Logistic regression assumes certain statistical patterns in the data and may slightly over- or under-represent the actual probability of admission of an applicant to that year’s class if those patterns were not uniformly present. For instance, this model might overestimate the probability for admittance for applicants just under the lowest score that a school accepted in 2009, and it might underestimate the probability for admittance for applicants just above that score.

The results of searches on this data may or may not reflect current admission probabilities at a given law school for 2011 admission.

In the reporting of the data by the law schools, an average LSAT score was used for candidates who had multiple scores.

Law schools consider many factors in the admission process other than UGPA and LSAT score, including letters of recommendation, work experience, personal statements, extracurricular and civic activities, diversity of classes, and many others.

Most applicants apply to a number of law schools based on a range of admission possibilities and other criteria. This search is designed to help you identify a number of schools, based on overlapping ranges, to explore further using the other searches in this Official Guide and additional research. Links to each law school are provided for additional information.

[2014]

The LSAC one gives me a higher chance at NYU than WUSTL as a high LSAT low GPA splitter...