Why not just use the raw scores? (Now: Why not scale 0-100?)

User avatar
Fiero85
Posts: 1909
Joined: Sun Jun 30, 2013 3:38 am

Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby Fiero85 » Mon Dec 09, 2013 4:43 pm

Not sure if anyone has addressed this previously, couldn't find it via search:

Why is the LSAT scaled to 120-180? The test is already 100 or so questions, so why not just use the raw scores? 0-100 is a pretty darn familiar scale to everyone.

Other than our "participation trophy" society where it isn't PC to give anyone a zero or even an F score on a test, I can't think of a good reason why LSAT scores are scaled at all.
Last edited by Fiero85 on Mon Dec 09, 2013 5:17 pm, edited 1 time in total.

Pancakes12
Posts: 269
Joined: Tue Jul 02, 2013 10:13 am

Re: Why not just use the raw scores?

Postby Pancakes12 » Mon Dec 09, 2013 4:47 pm

Some scores are harder to achieve on different tests
Last edited by Pancakes12 on Mon Dec 09, 2013 4:48 pm, edited 1 time in total.

User avatar
JuTMSY4
Posts: 265
Joined: Sat Jun 27, 2009 3:40 pm

Re: Why not just use the raw scores?

Postby JuTMSY4 » Mon Dec 09, 2013 4:47 pm

Fiero85 wrote:Not sure if anyone has addressed this previously, couldn't find it via search:

Why is the LSAT scaled to 120-180? The test is already 100 or so questions, so why not just use the raw scores? 0-100 is a pretty darn familiar scale to everyone.

Other than our "participation trophy" society where it isn't PC to give anyone a zero or even an F score on a test, I can't think of a good reason why LSAT scores are scaled at all.


Because some administrations of the LSAT are harder than others. Some are also easier.

User avatar
lawschool22
Posts: 3875
Joined: Thu Jul 25, 2013 5:47 pm

Re: Why not just use the raw scores?

Postby lawschool22 » Mon Dec 09, 2013 4:49 pm

Fiero85 wrote:Not sure if anyone has addressed this previously, couldn't find it via search:

Why is the LSAT scaled to 120-180? The test is already 100 or so questions, so why not just use the raw scores? 0-100 is a pretty darn familiar scale to everyone.

Other than our "participation trophy" society where it isn't PC to give anyone a zero or even an F score on a test, I can't think of a good reason why LSAT scores are scaled at all.


Because not every test has the same number of questions. Also, different tests are different levels of difficulty, so they have to curve it which requires a conversion. Finally they want the scores to fall along a certain distribution, so they scale the raw scores to fit that distribution. For instance because there are many more "easy" questions, in the lower raw score ranges they may it harder to move up a point in scaled score than they do as you progress up the scale. Presumably to score a 50 vs 60 raw score is easier than it is to score a 98 vs a 99, because likely to get that 99 you correctly answered a hard question.

ETA: Scooped sort of :D.

User avatar
Fiero85
Posts: 1909
Joined: Sun Jun 30, 2013 3:38 am

Re: Why not just use the raw scores?

Postby Fiero85 » Mon Dec 09, 2013 4:50 pm

Pancakes12 wrote:Some scores are harder to achieve on different tests


Ah yes that is true, but curving doesn't explain 120-180. I'm pretty sure they could just as easily curve it to make a "90" more consistent over time. I suppose that makes my question "why this scale vs. 0-100" rather than why not use raw scores. sorry for the confusion

Pancakes12
Posts: 269
Joined: Tue Jul 02, 2013 10:13 am

Re: Why not just use the raw scores?

Postby Pancakes12 » Mon Dec 09, 2013 4:52 pm

Fiero85 wrote:
Pancakes12 wrote:Some scores are harder to achieve on different tests


Ah yes that is true, but curving doesn't explain 120-180. I'm pretty sure they could just as easily curve it to make a "90" more consistent over time. I suppose that makes my question "why this scale vs. 0-100" rather than why not use raw scores. sorry for the confusion


Why not the 120-180 scale

User avatar
lawschool22
Posts: 3875
Joined: Thu Jul 25, 2013 5:47 pm

Re: Why not just use the raw scores?

Postby lawschool22 » Mon Dec 09, 2013 4:53 pm

http://www.powerscore.com/lsat/help/scale.cfm

Helpful info, although they don't fully address the "why 120-180" question. It was probably arbitrary.

User avatar
Fiero85
Posts: 1909
Joined: Sun Jun 30, 2013 3:38 am

Re: Why not just use the raw scores?

Postby Fiero85 » Mon Dec 09, 2013 4:55 pm

lawschool22 wrote:
Fiero85 wrote:Not sure if anyone has addressed this previously, couldn't find it via search:

Why is the LSAT scaled to 120-180? The test is already 100 or so questions, so why not just use the raw scores? 0-100 is a pretty darn familiar scale to everyone.

Other than our "participation trophy" society where it isn't PC to give anyone a zero or even an F score on a test, I can't think of a good reason why LSAT scores are scaled at all.


Because not every test has the same number of questions. Also, different tests are different levels of difficulty, so they have to curve it which requires a conversion. Finally they want the scores to fall along a certain distribution, so they scale the raw scores to fit that distribution. For instance because there are many more "easy" questions, in the lower raw score ranges they may it harder to move up a point in scaled score than they do as you progress up the scale. Presumably to score a 50 vs 60 raw score is easier than it is to score a 98 vs a 99, because likely to get that 99 you correctly answered a hard question.

ETA: Scooped sort of :D.


The first part is an easy fix, right? Every LSAT could be 100 questions if they so desired.

Secondly, yes, I wasn't factoring in the difficulty difference. But what about just curving the 0-100 scale?

And I think the easy vs. hard question distribution bands part doesn't make a difference. It is known now that a 150 to 160 jump is easier than 170 to 180. I believe the same could still work from 60-70 and 90-100.

User avatar
lawschool22
Posts: 3875
Joined: Thu Jul 25, 2013 5:47 pm

Re: Why not just use the raw scores?

Postby lawschool22 » Mon Dec 09, 2013 4:57 pm

And it looks like they do scale it to fit approximately a bell curve. I think the confusion people have is with the concept that the scale is pre-determined.

User avatar
lawschool22
Posts: 3875
Joined: Thu Jul 25, 2013 5:47 pm

Re: Why not just use the raw scores?

Postby lawschool22 » Mon Dec 09, 2013 4:57 pm

Fiero85 wrote:
lawschool22 wrote:
Fiero85 wrote:Not sure if anyone has addressed this previously, couldn't find it via search:

Why is the LSAT scaled to 120-180? The test is already 100 or so questions, so why not just use the raw scores? 0-100 is a pretty darn familiar scale to everyone.

Other than our "participation trophy" society where it isn't PC to give anyone a zero or even an F score on a test, I can't think of a good reason why LSAT scores are scaled at all.


Because not every test has the same number of questions. Also, different tests are different levels of difficulty, so they have to curve it which requires a conversion. Finally they want the scores to fall along a certain distribution, so they scale the raw scores to fit that distribution. For instance because there are many more "easy" questions, in the lower raw score ranges they may it harder to move up a point in scaled score than they do as you progress up the scale. Presumably to score a 50 vs 60 raw score is easier than it is to score a 98 vs a 99, because likely to get that 99 you correctly answered a hard question.

ETA: Scooped sort of :D.


The first part is an easy fix, right? Every LSAT could be 100 questions if they so desired.

Secondly, yes, I wasn't factoring in the difficulty difference. But what about just curving the 0-100 scale?

And I think the easy vs. hard question distribution bands part doesn't make a difference. It is known now that a 150 to 160 jump is easier than 170 to 180. I believe the same could still work from 60-70 and 90-100.


Yeah - my response was to your original question of why do they scale and not just use raw scores, not your second question of why specifically the actual numbers 120 to 180, which I do not know the answer to.
Last edited by lawschool22 on Mon Dec 09, 2013 4:58 pm, edited 1 time in total.

User avatar
Bildungsroman
Posts: 5548
Joined: Sun Apr 11, 2010 2:42 pm

Re: Why not just use the raw scores?

Postby Bildungsroman » Mon Dec 09, 2013 4:58 pm

Fiero85 wrote:
Pancakes12 wrote:Some scores are harder to achieve on different tests


Ah yes that is true, but curving doesn't explain 120-180. I'm pretty sure they could just as easily curve it to make a "90" more consistent over time. I suppose that makes my question "why this scale vs. 0-100" rather than why not use raw scores. sorry for the confusion


There are several reasons that jump to mind why they might scale to a different range than 0-100.

1. Scaling to a range that is close to the total number of questions doesn't adequately distinguish between raw and scaled scores.
2. They want to avoid prejudices, assumptions, associations, etc. that come with the 0-100 scale. You don't want people to import their prior perceptions about particular scores on the 100-point scale to the LSAT when the same numerical score should be interpreted differently between the two.
3. Why not use 120-180? 0-100 offers no benefits and plenty of detriments (e.g. the fact that someone will see 89 and think "B+" rather than, say, "98th percentile") and there's nothing innately great about a 0-100 scale except for a primitive obsession with round numbers.

User avatar
Fiero85
Posts: 1909
Joined: Sun Jun 30, 2013 3:38 am

Re: Why not just use the raw scores?

Postby Fiero85 » Mon Dec 09, 2013 4:59 pm

Pancakes12 wrote:
Fiero85 wrote:
Pancakes12 wrote:Some scores are harder to achieve on different tests


Ah yes that is true, but curving doesn't explain 120-180. I'm pretty sure they could just as easily curve it to make a "90" more consistent over time. I suppose that makes my question "why this scale vs. 0-100" rather than why not use raw scores. sorry for the confusion


Why not the 120-180 scale


I think that's the easy part...The test is already out of 100 or 101. Maybe I'm wrong, but I assume most would agree that 120-180 isn't the most obvious next step. Not that I bear any ill will towards the current scale, I'm just throwing it out there for discussion. Maybe they (correctly) think calling things online 180 sounds cooler than 100. haha

User avatar
Clearly
Posts: 4165
Joined: Sat Feb 11, 2012 4:09 pm

Re: Why not just use the raw scores?

Postby Clearly » Mon Dec 09, 2013 5:00 pm

My complete guess would be 180 developed mathematically as part of the system they first picked to equate the test. Perhaps each difficulty was assigned some point value, and the total point value of getting each question right was 180. Its kind of a chicken or the egg situation though, we likely won't ever know if the curve was designed around 180, or if 180 was a result of the curve.

User avatar
Fiero85
Posts: 1909
Joined: Sun Jun 30, 2013 3:38 am

Re: Why not just use the raw scores?

Postby Fiero85 » Mon Dec 09, 2013 5:01 pm

lawschool22 wrote:
Fiero85 wrote:
lawschool22 wrote:
Fiero85 wrote:Not sure if anyone has addressed this previously, couldn't find it via search:

Why is the LSAT scaled to 120-180? The test is already 100 or so questions, so why not just use the raw scores? 0-100 is a pretty darn familiar scale to everyone.

Other than our "participation trophy" society where it isn't PC to give anyone a zero or even an F score on a test, I can't think of a good reason why LSAT scores are scaled at all.


Because not every test has the same number of questions. Also, different tests are different levels of difficulty, so they have to curve it which requires a conversion. Finally they want the scores to fall along a certain distribution, so they scale the raw scores to fit that distribution. For instance because there are many more "easy" questions, in the lower raw score ranges they may it harder to move up a point in scaled score than they do as you progress up the scale. Presumably to score a 50 vs 60 raw score is easier than it is to score a 98 vs a 99, because likely to get that 99 you correctly answered a hard question.

ETA: Scooped sort of :D.


The first part is an easy fix, right? Every LSAT could be 100 questions if they so desired.

Secondly, yes, I wasn't factoring in the difficulty difference. But what about just curving the 0-100 scale?

And I think the easy vs. hard question distribution bands part doesn't make a difference. It is known now that a 150 to 160 jump is easier than 170 to 180. I believe the same could still work from 60-70 and 90-100.


Yeah - my response was to your original question of why do they scale and not just use raw scores, not your second question of why specifically the actual numbers 120 to 180, which I do not know the answer to.


Yeah sorry for switching up the question at hand guys, I'm just spitballing here. Thanks for the replies.

User avatar
Fiero85
Posts: 1909
Joined: Sun Jun 30, 2013 3:38 am

Re: Why not just use the raw scores?

Postby Fiero85 » Mon Dec 09, 2013 5:04 pm

Bildungsroman wrote:
Fiero85 wrote:
Pancakes12 wrote:Some scores are harder to achieve on different tests


Ah yes that is true, but curving doesn't explain 120-180. I'm pretty sure they could just as easily curve it to make a "90" more consistent over time. I suppose that makes my question "why this scale vs. 0-100" rather than why not use raw scores. sorry for the confusion


There are several reasons that jump to mind why they might scale to a different range than 0-100.

1. Scaling to a range that is close to the total number of questions doesn't adequately distinguish between raw and scaled scores.
2. They want to avoid prejudices, assumptions, associations, etc. that come with the 0-100 scale. You don't want people to import their prior perceptions about particular scores on the 100-point scale to the LSAT when the same numerical score should be interpreted differently between the two.
3. Why not use 120-180? 0-100 offers no benefits and plenty of detriments (e.g. the fact that someone will see 89 and think "B+" rather than, say, "98th percentile") and there's nothing innately great about a 0-100 scale except for a primitive obsession with round numbers.


These are good. Thank you!

User avatar
Fiero85
Posts: 1909
Joined: Sun Jun 30, 2013 3:38 am

Re: Why not just use the raw scores?

Postby Fiero85 » Mon Dec 09, 2013 5:08 pm

Clearly wrote:My complete guess would be 180 developed mathematically as part of the system they first picked to equate the test. Perhaps each difficulty was assigned some point value, and the total point value of getting each question right was 180. Its kind of a chicken or the egg situation though, we likely won't ever know if the curve was designed around 180, or if 180 was a result of the curve.


Interesting. I have no knowledge about the history of it myself, but that's why I asked. Thanks

User avatar
retaking23
Posts: 453
Joined: Thu Nov 01, 2012 10:34 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby retaking23 » Mon Dec 09, 2013 6:32 pm

It's a standardized test. That's why.

Edit: Sorry for the curt response, but like others have written, a score of x right on one exam does not necessarily translate to a score of x right on another exam. Of the many gripes I have with the LSAT, that it is standardized is not one of them.

User avatar
Jeffort
Posts: 1896
Joined: Wed Jun 18, 2008 4:43 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby Jeffort » Mon Dec 09, 2013 9:40 pm

Maybe it's because 180 is a cool number for many reasons:
http://en.wikipedia.org/wiki/180_%28number%29

My favorite reasons are:
The maximum possible score in one turn at darts (three triple 20s)
In Mario Kart Wii published by Nintendo, all battle mode games have a time limit of 180 seconds.
Each round of Boggle lasts 180 seconds in which players can make as many words as possible.


In mathematics

180 is an abundant number, with its proper divisors summing up to 366.[1][2] 180 is also a highly composite number, a positive integer with more divisors than any smaller positive integer. One of the consequences of 180 having so many divisors is that it is a practical number, meaning that any positive number smaller than 180 that is not a divisor of 180 can be expressed as the sum of some of 180's divisors.

180 is the sum of two square numbers: 122 + 62. It can be expressed as either the sum of six consecutive prime numbers: 19 + 23 + 29 + 31 + 37 + 41, or the sum of eight consecutive prime numbers: 11 + 13 + 17 + 19 + 23 + 29 + 31 + 37. 180 is an Ulam number, which can be expressed as a sum of earlier terms in the Ulam sequence only as 177 + 3.

180 is a 61-gonal number.[2]

Half a circle has 180 degrees.[3]

Summing Euler's totient function φ(x) over the first + 24 integers gives 180.

180 is a Harshad number in base 10, and in binary it is a digitally balanced number, since its binary representation has the same number of zeros as ones (10110100).
In sports

A "180" is a trick in extreme sports where a rider rotates half a turn while airborne and lands.
The maximum possible score in one turn at darts (three triple 20s).
In archery the gent's clout shooting distance is 180 yards.[4]

In other fields

180 is also:

The year AD 180 or 180 BC
180 is an astrological aspect considered to be negative in influence.[citation needed]
Scores on the Law School Admission Test (LSAT) are converted to a scaled score from 120 - 180.
Lithol Rubine BK, food dye classification #E180
In the Final Destination series, number 180 is the number of Death.
In Mario Kart Wii published by Nintendo, all battle mode games have a time limit of 180 seconds.
Each round of Boggle lasts 180 seconds in which players can make as many words as possible.

User avatar
USAO-vet
Posts: 53
Joined: Thu May 26, 2011 6:52 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby USAO-vet » Mon Dec 09, 2013 9:57 pm

The real question is, why not give you the fucking raw score immediately, pending the actual score being curved? At least that way one would have a decent perspective on how they've done. I mean, fuck, it's a god damn Scranton -- run that shit and give me some feedback.

User avatar
Nova
Posts: 9116
Joined: Sun Apr 15, 2012 8:55 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby Nova » Mon Dec 09, 2013 10:07 pm

USAO-vet wrote:The real question is, why not give you the fucking raw score immediately, pending the actual score being curved? At least that way one would have a decent perspective on how they've done. I mean, fuck, it's a god damn Scranton -- run that shit and give me some feedback.

the lsat isn't curved

its equated before the test is administered

http://www.top-law-schools.com/forums/v ... 0#p5497692

User avatar
ScottRiqui
Posts: 3640
Joined: Mon Nov 29, 2010 8:09 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby ScottRiqui » Mon Dec 09, 2013 10:10 pm

USAO-vet wrote:The real question is, why not give you the fucking raw score immediately, pending the actual score being curved? At least that way one would have a decent perspective on how they've done. I mean, fuck, it's a god damn Scranton -- run that shit and give me some feedback.


Well, they can't give them *immediately*, because not everyone finishes the test at the same time, and with a high-enough raw score in hand, you would essentially know all the answers to the questions you could remember. But I agree that they should be able to get you a tentative raw score faster than three weeks.

User avatar
lawschool22
Posts: 3875
Joined: Thu Jul 25, 2013 5:47 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby lawschool22 » Mon Dec 09, 2013 10:10 pm

USAO-vet wrote:The real question is, why not give you the fucking raw score immediately, pending the actual score being curved? At least that way one would have a decent perspective on how they've done. I mean, fuck, it's a god damn Scranton -- run that shit and give me some feedback.


They spend time ensuring there were no irregularities, administering any make up tests due to weather, conduct some hand scoring, handle any complaints, make sure all questions were good and shouldn't be removed from scoring, etc. before releasing scores. The time is not due to "curving" because as Nova rightly points out they don't curve it. It is "equated" prior to administration.

User avatar
USAO-vet
Posts: 53
Joined: Thu May 26, 2011 6:52 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby USAO-vet » Mon Dec 09, 2013 10:12 pm

Nova wrote:
USAO-vet wrote:The real question is, why not give you the fucking raw score immediately, pending the actual score being curved? At least that way one would have a decent perspective on how they've done. I mean, fuck, it's a god damn Scranton -- run that shit and give me some feedback.

the lsat isn't curved

its equated before the test is administered


Equated based on what, LSAC's guess? How about the questions that get tossed? I find it pretty fucking difficult to believe that they don't analyze difficultly and determine a curve based about how the actual test takers do? If true, that's quite retarded.

User avatar
Nova
Posts: 9116
Joined: Sun Apr 15, 2012 8:55 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby Nova » Mon Dec 09, 2013 10:14 pm

USAO-vet wrote:
Nova wrote:
USAO-vet wrote:The real question is, why not give you the fucking raw score immediately, pending the actual score being curved? At least that way one would have a decent perspective on how they've done. I mean, fuck, it's a god damn Scranton -- run that shit and give me some feedback.

the lsat isn't curved

its equated before the test is administered


Equated based on what, LSAC's guess? How about the questions that get tossed? I find it pretty fucking difficult to believe that they don't analyze difficultly and determine a curve based about how the actual test takers do? If true, that's quite retarded.

see SOJ's post in the link i posted. Its not retarded because they already know how difficult the test is based on prior administrations. they can still tweak the equated scores after the administration if something seems really off.

User avatar
lawschool22
Posts: 3875
Joined: Thu Jul 25, 2013 5:47 pm

Re: Why not just use the raw scores? (Now: Why not scale 0-100?)

Postby lawschool22 » Mon Dec 09, 2013 10:15 pm

USAO-vet wrote:
Nova wrote:
USAO-vet wrote:The real question is, why not give you the fucking raw score immediately, pending the actual score being curved? At least that way one would have a decent perspective on how they've done. I mean, fuck, it's a god damn Scranton -- run that shit and give me some feedback.

the lsat isn't curved

its equated before the test is administered


Equated based on what, LSAC's guess? How about the questions that get tossed? I find it pretty fucking difficult to believe that they don't analyze difficultly and determine a curve based about how the actual test takers do? If true, that's quite retarded.


They don't. There is an experimental section for every administration and based on the results of that, they determine the relative difficulty of that question (which is an oversimplification). Then by the makeup of the difficulty of the questions (which was pre-determined since these questions have already been tested) they can equate the test and determine the scale. This is why there is variability in the percentiles between administrations.

And it's not retarded because they don't want your score to be skewed by the relative ability of the test takers in a given administration. Just because you happened to take the test with a group of really smart people doesn't mean you don't deserve a 172 or whatever. It's designed to test your true ability, irrespective of how well your particular group did.




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: dontsaywhatyoumean, potatocowpower and 1 guest