Curve for October 2010?

User avatar
pppokerface
Posts: 205
Joined: Tue Oct 12, 2010 4:45 pm

Re: Curve for October 2010?

Postby pppokerface » Fri Oct 15, 2010 1:27 pm

Funnily enough, I liked the June games more and had an easier time.

The October games were NOT tough-the games themselves. What made them hard was that they looked simple, but they were time consuming. By the time a lot of people got to the final game, they had little time. Or at least that is what happened to me. The curve is determined by how people did on the experimentals on the previous tests and how they did on the real thing...? from what I am reading. Well it may be better than friggin -10 which has not happened in awhile-I wouldnt be surprised if time due to lack of deductions hurt a lot of people. Plus TLS always underestimates the curve.

User avatar
incompetentia
Posts: 2307
Joined: Thu Sep 30, 2010 2:57 pm

Re: Curve for October 2010?

Postby incompetentia » Fri Oct 15, 2010 2:11 pm

Remember that not all -10 curves are created equal. Just because the general public finds something more difficult does not mean the curve at 170 will be affected

Sandro
Posts: 2526
Joined: Sat Jul 18, 2009 12:12 am

Re: Curve for October 2010?

Postby Sandro » Fri Oct 15, 2010 2:19 pm

incompetentia wrote:Remember that not all -10 curves are created equal. Just because the general public finds something more difficult does not mean the curve at 170 will be affected


Im probably completely wrong but I thought sections of the test were created by pulling together questions from experimental sections, with the questions trying to be equated from test to test to be the same level of difficulty overall. Wouldn't it make sense that questions missed by a greater % of ppl would = harder questions ??

User avatar
kkklick
Posts: 1021
Joined: Mon Jun 07, 2010 8:33 pm

Re: Curve for October 2010?

Postby kkklick » Fri Oct 15, 2010 2:21 pm

incompetentia wrote:Remember that not all -10 curves are created equal. Just because the general public finds something more difficult does not mean the curve at 170 will be affected

So what is your best educated guess?

User avatar
incompetentia
Posts: 2307
Joined: Thu Sep 30, 2010 2:57 pm

Re: Curve for October 2010?

Postby incompetentia » Fri Oct 15, 2010 2:59 pm

Sandro777 wrote:
inck wrote:Remember that not all -10 curves are created equal. Just because the general public finds something more difficult does not mean the curve at 170 will be affected


Im probably completely wrong but I thought sections of the test were created by pulling together questions from experimental sections, with the questions trying to be equated from test to test to be the same level of difficulty overall. Wouldn't it make sense that questions missed by a greater % of ppl would = harder questions ??

These two are not incompatible. Maybe normally you have 2% of people missing less than 10 and 50% of people missing less than 40, but if you have more difficult questions that the top people STILL get, then you might still have 2% of people missing less than 10 and maybe only 25% of people missing less than 40.

The top scorers by definition have more correct questions than the lower scorers, but the question is by how much. This is where the curve can deviate from itself, for lack of a better description.
Last edited by incompetentia on Sat Aug 27, 2011 8:57 pm, edited 1 time in total.

User avatar
DrackedaryMaster
Posts: 181
Joined: Sun Jul 04, 2010 5:11 pm

Re: Curve for October 2010?

Postby DrackedaryMaster » Fri Oct 15, 2010 5:24 pm

JJDancer wrote:
DrackedaryMaster wrote:This may not mean anything in the end, and I'm sure someone else probably already pointed this out but at least it gives me somehing to do while I battle PT withdrawl syndrome. At least I hope the raw score pattern holds up.

Okay, after looking at the raw score charts, I noticed some very consistent data between the June and October LSATs and the magical 170/160 thresholds. Between the two tests, the raw score needed to get 170 on the October test has not been higher than (1) more correct answer choice in over 16 YEARS from the June test. Most of the time, the raw score requirements either stayed the same or went down for the October test. In the rare cases it went up, the lowest raw score required to hit 170/160 never went up by more than 1, except in 1992 and 1994 for 170 hopefuls and 1999 for 160 hopefuls) December, on the other hand has some pretty wild swings. But it looks like the pattern for June and Sept/Oct has held consistent except for the noted exceptions.

Based on this, I think it is somewhat probable that the raw score required to hit 170 will be no higher than 88. This means the curve should be no harsher than -13, or (-12 if a question gets thrown out). Otherwise, it will be the first time in 16 years that this raw score trend between June and September/October will have been broken.

Maybe I'm not understanding you correctly but Sept 2009 required 90/101 for a 170. Same with June 2009. Oct 2008 required 91/100 for a 170.


My point is that there is some major statistical consistency between the Raw Scores required to hit 170/160 on the June and September/October LSAT. This doesn’t work for any other LSAT test combination, only those two:

Looking at every June and September/October LSAT from 1991 to the present, take a look at the difference in the Raw Score required to hit 170 on the latter test as compared to the former. You should notice something stick out. With the exception of 1992 & 1994, the Raw Score required to get a 170 on the September/October LSAT has been, at most, no more than 1 more correct answer than what was required to get 170 on the June test. 1994 was the last time it took 2 more correct answer choices than June to score. And 1992 was the only time it took more than 2 correct answer choices to hit the 170 mark in Sept/Oct as compared to June.

If you look at December’s Raw Scores though, they are all over the place, same as June when the beginning of the testing year starts to get a feel of where LSAC is going. My point is, the September/October LSAT does not appear to be a place where wild raw score fluctuations occur, unless it actually LOWERS the required score, as evidenced by the statistics.

So for those of you thinking the curve will be -11, keep this in mind. Assuming no questions get booted out, the required Raw Score to get 170 would be 90, which would be 3 more correct answers than required in June. As evidenced above, that wide an up-tick has not occurred between these two tests since 1992!

If 87 was the required raw score for June, statistics hold that it would be extremely rare for it to go up by 2 or more correct answers on the October Test. Statistics lead to one of the following scenarios playing out:

87 required for 170 on October 10 (-14 curve) or even a lower raw score and curve
87 required for 170 on October 10 with 1 Question Booted out (-13 curve)
88 required for 170 on October 10 (-13 curve)
88 required for 170 on October 10 with 1 Question Booted out (-12 curve)

Anything else would be highly unusual.

User avatar
Ragged
Posts: 1509
Joined: Wed Oct 21, 2009 12:39 pm

Re: Curve for October 2010?

Postby Ragged » Fri Oct 15, 2010 5:29 pm

DrackedaryMaster wrote:
JJDancer wrote:
DrackedaryMaster wrote:This may not mean anything in the end, and I'm sure someone else probably already pointed this out but at least it gives me somehing to do while I battle PT withdrawl syndrome. At least I hope the raw score pattern holds up.

Okay, after looking at the raw score charts, I noticed some very consistent data between the June and October LSATs and the magical 170/160 thresholds. Between the two tests, the raw score needed to get 170 on the October test has not been higher than (1) more correct answer choice in over 16 YEARS from the June test. Most of the time, the raw score requirements either stayed the same or went down for the October test. In the rare cases it went up, the lowest raw score required to hit 170/160 never went up by more than 1, except in 1992 and 1994 for 170 hopefuls and 1999 for 160 hopefuls) December, on the other hand has some pretty wild swings. But it looks like the pattern for June and Sept/Oct has held consistent except for the noted exceptions.

Based on this, I think it is somewhat probable that the raw score required to hit 170 will be no higher than 88. This means the curve should be no harsher than -13, or (-12 if a question gets thrown out). Otherwise, it will be the first time in 16 years that this raw score trend between June and September/October will have been broken.

Maybe I'm not understanding you correctly but Sept 2009 required 90/101 for a 170. Same with June 2009. Oct 2008 required 91/100 for a 170.


My point is that there is some major statistical consistency between the Raw Scores required to hit 170/160 on the June and September/October LSAT. This doesn’t work for any other LSAT test combination, only those two:

Looking at every June and September/October LSAT from 1991 to the present, take a look at the difference in the Raw Score required to hit 170 on the latter test as compared to the former. You should notice something stick out. With the exception of 1992 & 1994, the Raw Score required to get a 170 on the September/October LSAT has been, at most, no more than 1 more correct answer than what was required to get 170 on the June test. 1994 was the last time it took 2 more correct answer choices than June to score. And 1992 was the only time it took more than 2 correct answer choices to hit the 170 mark in Sept/Oct as compared to June.

If you look at December’s Raw Scores though, they are all over the place, same as June when the beginning of the testing year starts to get a feel of where LSAC is going. My point is, the September/October LSAT does not appear to be a place where wild raw score fluctuations occur, unless it actually LOWERS the required score, as evidenced by the statistics.

So for those of you thinking the curve will be -11, keep this in mind. Assuming no questions get booted out, the required Raw Score to get 170 would be 90, which would be 3 more correct answers than required in June. As evidenced above, that wide an up-tick has not occurred between these two tests since 1992!

If 87 was the required raw score for June, statistics hold that it would be extremely rare for it to go up by 2 or more correct answers on the October Test. Statistics lead to one of the following scenarios playing out:

87 required for 170 on October 10 (-14 curve) or even a lower raw score and curve
87 required for 170 on October 10 with 1 Question Booted out (-13 curve)
88 required for 170 on October 10 (-13 curve)
88 required for 170 on October 10 with 1 Question Booted out (-12 curve)

Anything else would be highly unusual.



We all hope you're right.

User avatar
s0ph1e2007
Posts: 1054
Joined: Sat Oct 10, 2009 10:37 pm

Re: Curve for October 2010?

Postby s0ph1e2007 » Fri Oct 15, 2010 5:31 pm

I mean I'd love for it to be -13, but that seems very very very unlikely right now.
I'm in the hope: -12, prediction: -11
range right now....
although really really wouldn't be surprised if it were -10

User avatar
catsparka
Posts: 267
Joined: Fri Sep 10, 2010 4:50 pm

Re: Curve for October 2010?

Postby catsparka » Fri Oct 15, 2010 5:34 pm

DrackedaryMaster wrote:
JJDancer wrote:
DrackedaryMaster wrote:This may not mean anything in the end, and I'm sure someone else probably already pointed this out but at least it gives me somehing to do while I battle PT withdrawl syndrome. At least I hope the raw score pattern holds up.

Okay, after looking at the raw score charts, I noticed some very consistent data between the June and October LSATs and the magical 170/160 thresholds. Between the two tests, the raw score needed to get 170 on the October test has not been higher than (1) more correct answer choice in over 16 YEARS from the June test. Most of the time, the raw score requirements either stayed the same or went down for the October test. In the rare cases it went up, the lowest raw score required to hit 170/160 never went up by more than 1, except in 1992 and 1994 for 170 hopefuls and 1999 for 160 hopefuls) December, on the other hand has some pretty wild swings. But it looks like the pattern for June and Sept/Oct has held consistent except for the noted exceptions.

Based on this, I think it is somewhat probable that the raw score required to hit 170 will be no higher than 88. This means the curve should be no harsher than -13, or (-12 if a question gets thrown out). Otherwise, it will be the first time in 16 years that this raw score trend between June and September/October will have been broken.

Maybe I'm not understanding you correctly but Sept 2009 required 90/101 for a 170. Same with June 2009. Oct 2008 required 91/100 for a 170.


My point is that there is some major statistical consistency between the Raw Scores required to hit 170/160 on the June and September/October LSAT. This doesn’t work for any other LSAT test combination, only those two:

Looking at every June and September/October LSAT from 1991 to the present, take a look at the difference in the Raw Score required to hit 170 on the latter test as compared to the former. You should notice something stick out. With the exception of 1992 & 1994, the Raw Score required to get a 170 on the September/October LSAT has been, at most, no more than 1 more correct answer than what was required to get 170 on the June test. 1994 was the last time it took 2 more correct answer choices than June to score. And 1992 was the only time it took more than 2 correct answer choices to hit the 170 mark in Sept/Oct as compared to June.

If you look at December’s Raw Scores though, they are all over the place, same as June when the beginning of the testing year starts to get a feel of where LSAC is going. My point is, the September/October LSAT does not appear to be a place where wild raw score fluctuations occur, unless it actually LOWERS the required score, as evidenced by the statistics.

So for those of you thinking the curve will be -11, keep this in mind. Assuming no questions get booted out, the required Raw Score to get 170 would be 90, which would be 3 more correct answers than required in June. As evidenced above, that wide an up-tick has not occurred between these two tests since 1992!

If 87 was the required raw score for June, statistics hold that it would be extremely rare for it to go up by 2 or more correct answers on the October Test. Statistics lead to one of the following scenarios playing out:

87 required for 170 on October 10 (-14 curve) or even a lower raw score and curve
87 required for 170 on October 10 with 1 Question Booted out (-13 curve)
88 required for 170 on October 10 (-13 curve)
88 required for 170 on October 10 with 1 Question Booted out (-12 curve)

Anything else would be highly unusual.


PLEASE

Sandro
Posts: 2526
Joined: Sat Jul 18, 2009 12:12 am

Re: Curve for October 2010?

Postby Sandro » Fri Oct 15, 2010 6:00 pm

a -12 or -13 curve would make me do the rain dance, no matter if I got a 170 or not...

nStiver
Posts: 388
Joined: Thu Nov 06, 2008 4:15 am

Re: Curve for October 2010?

Postby nStiver » Fri Oct 15, 2010 6:04 pm

-50

User avatar
3|ink
Posts: 7331
Joined: Wed Dec 16, 2009 5:23 pm

Re: Curve for October 2010?

Postby 3|ink » Fri Oct 15, 2010 6:48 pm

nStiver wrote:-50


-2

JJDancer
Posts: 1564
Joined: Sun Jul 26, 2009 7:41 pm

Re: Curve for October 2010?

Postby JJDancer » Fri Oct 15, 2010 7:21 pm

DrackedaryMaster wrote:My point is that there is some major statistical consistency between the Raw Scores required to hit 170/160 on the June and September/October LSAT. This doesn’t work for any other LSAT test combination, only those two:

Ah ok, gotcha. Interesting stuff. Is this regardless of whether June had 99/100/101 questions as compared to Oct?
For example: 90/100 to get 170 in June
89/101 to get 170 in Oct
It's one less raw score, but out of 1 more question so..

PS: Oct had 101 right? And June 2010 had 99?

User avatar
DrackedaryMaster
Posts: 181
Joined: Sun Jul 04, 2010 5:11 pm

Re: Curve for October 2010?

Postby DrackedaryMaster » Fri Oct 15, 2010 7:47 pm

There are some years where one test had 100, the other 101, or 99, but the pattern has held. But if no question gets booted out in Oct 2010, this would be the first instance of one test having two more questions than the other. I still think with the equating that that the pattern will hold. Of course, this is probably wishful thinking on my part. Still.

bartleby
Posts: 1315
Joined: Thu May 29, 2008 3:23 am

Re: Curve for October 2010?

Postby bartleby » Fri Oct 15, 2010 7:49 pm

This thread is giving me hope. I don't know how I'm going to last another two weeks. Or worse...Nov 2...

NJcollegestudent
Posts: 490
Joined: Sun Jun 06, 2010 5:12 pm

Re: Curve for October 2010?

Postby NJcollegestudent » Fri Oct 15, 2010 7:52 pm

DrackedaryMaster wrote:There are some years where one test had 100, the other 101, or 99, but the pattern has held. But if no question gets booted out in Oct 2010, this would be the first instance of one test having two more questions than the other. I still think with the equating that that the pattern will hold. Of course, this is probably wishful thinking on my part. Still.



The question then becomes, how many wrong for a 160 and if there is a correlation between June and October for that range.

User avatar
Lasers
Posts: 1576
Joined: Sat Jul 10, 2010 6:46 pm

Re: Curve for October 2010?

Postby Lasers » Fri Oct 15, 2010 8:17 pm

DrackedaryMaster wrote:
JJDancer wrote:
DrackedaryMaster wrote:This may not mean anything in the end, and I'm sure someone else probably already pointed this out but at least it gives me somehing to do while I battle PT withdrawl syndrome. At least I hope the raw score pattern holds up.

Okay, after looking at the raw score charts, I noticed some very consistent data between the June and October LSATs and the magical 170/160 thresholds. Between the two tests, the raw score needed to get 170 on the October test has not been higher than (1) more correct answer choice in over 16 YEARS from the June test. Most of the time, the raw score requirements either stayed the same or went down for the October test. In the rare cases it went up, the lowest raw score required to hit 170/160 never went up by more than 1, except in 1992 and 1994 for 170 hopefuls and 1999 for 160 hopefuls) December, on the other hand has some pretty wild swings. But it looks like the pattern for June and Sept/Oct has held consistent except for the noted exceptions.

Based on this, I think it is somewhat probable that the raw score required to hit 170 will be no higher than 88. This means the curve should be no harsher than -13, or (-12 if a question gets thrown out). Otherwise, it will be the first time in 16 years that this raw score trend between June and September/October will have been broken.

Maybe I'm not understanding you correctly but Sept 2009 required 90/101 for a 170. Same with June 2009. Oct 2008 required 91/100 for a 170.


My point is that there is some major statistical consistency between the Raw Scores required to hit 170/160 on the June and September/October LSAT. This doesn’t work for any other LSAT test combination, only those two:

Looking at every June and September/October LSAT from 1991 to the present, take a look at the difference in the Raw Score required to hit 170 on the latter test as compared to the former. You should notice something stick out. With the exception of 1992 & 1994, the Raw Score required to get a 170 on the September/October LSAT has been, at most, no more than 1 more correct answer than what was required to get 170 on the June test. 1994 was the last time it took 2 more correct answer choices than June to score. And 1992 was the only time it took more than 2 correct answer choices to hit the 170 mark in Sept/Oct as compared to June.

If you look at December’s Raw Scores though, they are all over the place, same as June when the beginning of the testing year starts to get a feel of where LSAC is going. My point is, the September/October LSAT does not appear to be a place where wild raw score fluctuations occur, unless it actually LOWERS the required score, as evidenced by the statistics.

So for those of you thinking the curve will be -11, keep this in mind. Assuming no questions get booted out, the required Raw Score to get 170 would be 90, which would be 3 more correct answers than required in June. As evidenced above, that wide an up-tick has not occurred between these two tests since 1992!

If 87 was the required raw score for June, statistics hold that it would be extremely rare for it to go up by 2 or more correct answers on the October Test. Statistics lead to one of the following scenarios playing out:

87 required for 170 on October 10 (-14 curve) or even a lower raw score and curve
87 required for 170 on October 10 with 1 Question Booted out (-13 curve)
88 required for 170 on October 10 (-13 curve)
88 required for 170 on October 10 with 1 Question Booted out (-12 curve)

Anything else would be highly unusual.

that all makes too much sense.

User avatar
incompetentia
Posts: 2307
Joined: Thu Sep 30, 2010 2:57 pm

Re: Curve for October 2010?

Postby incompetentia » Fri Oct 15, 2010 9:01 pm

3|ink wrote:
nStiver wrote:-50


-2

-250

pseudonym1
Posts: 89
Joined: Tue Feb 09, 2010 10:02 am

Re: Curve for October 2010?

Postby pseudonym1 » Tue Oct 19, 2010 6:14 pm

DrackedaryMaster wrote:There are some years where one test had 100, the other 101, or 99, but the pattern has held. But if no question gets booted out in Oct 2010, this would be the first instance of one test having two more questions than the other. I still think with the equating that that the pattern will hold. Of course, this is probably wishful thinking on my part. Still.


June 2010 = -12 Curve = 87 right
Oct. 2010 (if it had 101 questions), a -12 curve would still mean you need 89, -11 = 90, etc. etc.

You seem to be saying the curve will be -13, suggesting October was harder than June.

User avatar
incompetentia
Posts: 2307
Joined: Thu Sep 30, 2010 2:57 pm

Re: Curve for October 2010?

Postby incompetentia » Tue Oct 19, 2010 6:20 pm

Problem is that June 2010 was in and of itself an unusual test relative to the curve. It's possible that June will be seen as an anomaly and that we'll get -10 again

am060459
Posts: 609
Joined: Mon Jun 29, 2009 12:14 am

Re: Curve for October 2010?

Postby am060459 » Tue Oct 19, 2010 6:21 pm

-11 hoping for -12.


i really dont see a -10.

User avatar
DrackedaryMaster
Posts: 181
Joined: Sun Jul 04, 2010 5:11 pm

Re: Curve for October 2010?

Postby DrackedaryMaster » Tue Oct 19, 2010 6:30 pm

pseudonym1 wrote:
DrackedaryMaster wrote:There are some years where one test had 100, the other 101, or 99, but the pattern has held. But if no question gets booted out in Oct 2010, this would be the first instance of one test having two more questions than the other. I still think with the equating that that the pattern will hold. Of course, this is probably wishful thinking on my part. Still.


June 2010 = -12 Curve = 87 right
Oct. 2010 (if it had 101 questions), a -12 curve would still mean you need 89, -11 = 90, etc. etc.

You seem to be saying the curve will be -13, suggesting October was harder than June.



No, I think if the curve is -12, then 1 question is going to get booted out, reducing the test to 100 questions and keeping the pattern intact. I'm actually starting to believe it is more and more likely a question will get booted out. The stats say that "corrections" to the raw score occur in December and June. Whether I think October was easier/harder than June is irrelevant. All I know is that LSAC seems to have consistent data that has allowed them to keep these two tests closely equated raw score wise in terms of difficulty for the past 20 years. If no question gets thrown out, -13 or lower becomes very probable.

pseudonym1
Posts: 89
Joined: Tue Feb 09, 2010 10:02 am

Re: Curve for October 2010?

Postby pseudonym1 » Tue Oct 19, 2010 6:35 pm

They may remove 1 question from one of the LR sections, making the curve -12, thus being equal to June. That would be fair considering a lot of the retakers were fucked by June's test. On the other hand, a -11 would be a teabagging for those same people.

Destined
Posts: 65
Joined: Sun Aug 08, 2010 3:55 pm

Re: Curve for October 2010?

Postby Destined » Tue Oct 19, 2010 6:36 pm

i know this has been answered a gazillion times, but how many questions were on LG for October 2010?

User avatar
DrackedaryMaster
Posts: 181
Joined: Sun Jul 04, 2010 5:11 pm

Re: Curve for October 2010?

Postby DrackedaryMaster » Tue Oct 19, 2010 6:49 pm

incompetentia wrote:Problem is that June 2010 was in and of itself an unusual test relative to the curve. It's possible that June will be seen as an anomaly and that we'll get -10 again



Unusual how? The recent precedent that one could get a raw score as low as 87 and still get 170 was reestablished on PT59.

I think it has to do with the fact that the experimentals that eventually made up PT59-61 were all tested in a similar timeframe (2008-onward) when the economy was starting to tank and the rise in the number of LSAT takers rose. This had to have a significant effect on the raw score requirements. If PT59 had been an anomaly, then "normalcy" should have returned to PT60 at the beginning of LSAC's testing cycle. But it didn't. And I don't believe "normalcy" will return with PT61 either given that LSAC has been very consistent in equating the two tests. PT62 (Dec 2010) is an entirely different story. If they want to raise the raw score back up to 90, it will probably be there.

I could be wrong, but I sure hope I'm not.




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: jagerbom79, JoshLyman13, sd812, wildquest8200, Yahoo [Bot] and 9 guests