Curve for October 2010? Forum

[kkklick]

Damn, this curve is going to be much more interesting than I thought

[incompetentia]

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.

I see where you're coming from, but note that the only years where the NUMBER of questions increased were in 1992 (one of your 'outliers') and 2005 (raw for 170 steady). Even if a question gets tossed, the number of questions WILL increase.
Note that in these cases, there is no reason to expect that the raw score goes DOWN, and rather the evidence is slightly stronger that it goes UP.

[Blumpbeef]

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

PT59 had a satanic LR section.

Both 59 and 60 were pretty hard tests all around. 61 wan'st easy, but it also wasn't a cakewalk.

[djjf39]

61 wan'st easy, but it also wasn't a cakewalk.

Not sure what to make of this.

[ThreeYears]

61 wan'st easy, but it also wasn't a cakewalk.

Not sure what to make of this.

i guess the "mimicry vertigo" of cakewalk makes it extremely difficult?

[incompetentia]

61 wan'st easy, but it also wasn't a cakewalk.

Not sure what to make of this.

It didn't cater to all classes and races of people.
Shame on you, LSAT.

[djjf39]

61 wan'st easy, but it also wasn't a cakewalk.

Not sure what to make of this.

i guess the "mimicry vertigo" of cakewalk makes it extremely difficult?

Oh wow , I am surprised I repressed that RC passage so thoroughly. I though the statement was nonsensical, but I now see how severely traumatized June's LSAT left me.

[incompetentia]

Oh wow , I am surprised I repressed that RC passage so thoroughly. I though the statement was nonsensical, but I now see how severely traumatized June's LSAT left me.

There, there. I'm sure your positive-light sculptures will be a runaway success.

[Ragged]

61 wan'st easy, but it also wasn't a cakewalk.

Are you referring to that RC passage about the cakewalk? So you mean it wasn't easy, but it wasn't as hard as the test that featured that passage? Makes sense.

[DrackedaryMaster]

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

PT59 had a satanic LR section.

Both 59 and 60 were pretty hard tests all around. 61 wan'st easy, but it also wasn't a cakewalk.

If Satan's name is Noguchi, then I would agree. I thought the RC for PT59 was ridiculously hard and single-handedly set the curve for that test.

But let's assume you're right about the LR section being unusually difficult on PT59. I know that one of PT61's real LR sections was an experimental from PT59 due to using the TLS search function. So how do you think someone (a good scorer, say) would have handled the two LR's on PT59 plus the experimental LR with the brick-houses and/or Rule of Law and other traps set up? PT61's RC was also tested as the experimental on PT55. I feel bad for those that had to juggle both the Talk-Story and Transnationalism. Admittedly, they were probably more pi$$ed when the latter didn't count, but it wasn't all that great. And although information from the boards on Feb 2009 is sketchy, our LG apparently got tested with another somewhat difficult LG that counted on the Feb 2009 LSAT, with most relieved that PT61's LG was fake for them.

I guess my point is the composition of PT61 got experimented on tests that would have probably had a strong mental affect on the test takers. I think RC comes in slightly easier from what it was tested against on PT55 because all four passages were not ambiguous, but the questions were still challenging. If I had to guess, I'd say our LG was somewhat harder than what the Feb 2009 takers faced as their real LG, but this is based primarily on thread posts and is not really meaningful. The Dec 2009 testers with a PT61 LR Experimental, well that would be tough, with the new question stems and what-not in addition to the real LR sections. True, maybe they caught on it was experimental. But some of us on PT61 thought the same.

That's the environment the scale for our test was set to. For the record, I haven't found the other LR section at all yet. My guess is that it was an experimental on one of the Sabbath tests.

[dcman06]

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

23

There have consistently been 23 questions in games for at least the last five or so PT's.

[ThreeYears]

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

PT59 had a satanic LR section.

Both 59 and 60 were pretty hard tests all around. 61 wan'st easy, but it also wasn't a cakewalk.

If Satan's name is Noguchi, then I would agree. I thought the RC for PT59 was ridiculously hard and single-handedly set the curve for that test.

But let's assume you're right about the LR section being unusually difficult on PT59. I know that one of PT61's real LR sections was an experimental from PT59 due to using the TLS search function. So how do you think someone (a good scorer, say) would have handled the two LR's on PT59 plus the experimental LR with the brick-houses and/or Rule of Law and other traps set up? PT61's RC was also tested as the experimental on PT55. I feel bad for those that had to juggle both the Talk-Story and Transnationalism. Admittedly, they were probably more pi$$ed when the latter didn't count, but it wasn't all that great. And although information from the boards on Feb 2009 is sketchy, our LG apparently got tested with another somewhat difficult LG that counted on the Feb 2009 LSAT, with most relieved that PT61's LG was fake for them.

I guess my point is the composition of PT61 got experimented on tests that would have probably had a strong mental affect on the test takers. I think RC comes in slightly easier from what it was tested against on PT55 because all four passages were not ambiguous, but the questions were still challenging. If I had to guess, I'd say our LG was somewhat harder than what the Feb 2009 takers faced as their real LG, but this is based primarily on thread posts and is not really meaningful. The Dec 2009 testers with a PT61 LR Experimental, well that would be tough, with the new question stems and what-not in addition to the real LR sections. True, maybe they caught on it was experimental. But some of us on PT61 thought the same.

That's the environment the scale for our test was set to. For the record, I haven't found the other LR section at all yet. My guess is that it was an experimental on one of the Sabbath tests.

I want you to be on my research team.

[kkklick]

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

PT59 had a satanic LR section.

Both 59 and 60 were pretty hard tests all around. 61 wan'st easy, but it also wasn't a cakewalk.

If Satan's name is Noguchi, then I would agree. I thought the RC for PT59 was ridiculously hard and single-handedly set the curve for that test.

But let's assume you're right about the LR section being unusually difficult on PT59. I know that one of PT61's real LR sections was an experimental from PT59 due to using the TLS search function. So how do you think someone (a good scorer, say) would have handled the two LR's on PT59 plus the experimental LR with the brick-houses and/or Rule of Law and other traps set up? PT61's RC was also tested as the experimental on PT55. I feel bad for those that had to juggle both the Talk-Story and Transnationalism. Admittedly, they were probably more pi$$ed when the latter didn't count, but it wasn't all that great. And although information from the boards on Feb 2009 is sketchy, our LG apparently got tested with another somewhat difficult LG that counted on the Feb 2009 LSAT, with most relieved that PT61's LG was fake for them.

I guess my point is the composition of PT61 got experimented on tests that would have probably had a strong mental affect on the test takers. I think RC comes in slightly easier from what it was tested against on PT55 because all four passages were not ambiguous, but the questions were still challenging. If I had to guess, I'd say our LG was somewhat harder than what the Feb 2009 takers faced as their real LG, but this is based primarily on thread posts and is not really meaningful. The Dec 2009 testers with a PT61 LR Experimental, well that would be tough, with the new question stems and what-not in addition to the real LR sections. True, maybe they caught on it was experimental. But some of us on PT61 thought the same.

That's the environment the scale for our test was set to. For the record, I haven't found the other LR section at all yet. My guess is that it was an experimental on one of the Sabbath tests.

I want you to be on my research team.

+1

[DrackedaryMaster]

I want you to be on my research team.

No you don't. As much as I love numbers crunching, I had my a$$ handed to me trying to predict 2008 election returns. It's not that I had the Presidential contest in doubt, but the margins and areas where the totals came in that bothered me. To this day, Penn, Ohio, FL, and VA still bother me. It's not that all these states that went blue was a surprise, but the locations and margins from where Obama got his support. Data I had said that in order to win all four, he'd have to poll support higher than what most major liberals are able to garner. If you look at those four states, the most successful Democratic candidates usually are more to the center/right at the state/national level. It has been this way given the make-up of those state's populaces. The "conservative" or "centrist" Democrat was always able to draw the largest share (90%+) of party supporters and the best chance of drawing a majority of moderates, while the more liberal Democrat usually was about 7-10% behind in own party support (80%).

Most surprising was OH. Obama polled better in party support and in rural "blue" areas than Ted Strickland! Not far behind was PA where Obama bested the popular moderate Rendell, taking sizable majorities in Berks and Bucks Counties, and increasing his margins in Southwestern PA despite McCain's gains on the OH/PA border. Loudon County in VA was more acceptable, but still Obama outpolled support for Kaine, although Warner did best him. The I-4 corridor was also a shocker. That region tilts a lot more center-right than it's neutral reputation would suggest and Obama's margins were surprising. At one point, I thought he was going to win in Montana, which is a state no candidate classified as a "liberal" should have any business winning any more than a candidate in New York running as a "conservative" should. And while people were calling this a paradigm shift, just two years later, everything seems to be going back to the way it was pre-2008 given the 2010 projections for particular regions.

[ThreeYears]

I want you to be on my research team.

No you don't. As much as I love numbers crunching, I had my a$$ handed to me trying to predict 2008 election returns. It's not that I had the Presidential contest in doubt, but the margins and areas where the totals came in that bothered me. To this day, Penn, Ohio, FL, and VA still bother me. It's not that all these states that went blue was a surprise, but the locations and margins from where Obama got his support. Data I had said that in order to win all four, he'd have to poll support higher than what most major liberals are able to garner. If you look at those four states, the most successful Democratic candidates usually are more to the center/right at the state/national level. It has been this way given the make-up of those state's populaces. The "conservative" or "centrist" Democrat was always able to draw the largest share (90%+) of party supporters and the best chance of drawing a majority of moderates, while the more liberal Democrat usually was about 7-10% behind in own party support (80%).

Most surprising was OH. Obama polled better in party support and in rural "blue" areas than Ted Strickland! Not far behind was PA where Obama bested the popular moderate Rendell, taking sizable majorities in Berks and Bucks Counties, and increasing his margins in Southwestern PA despite McCain's gains on the OH/PA border. Loudon County in VA was more acceptable, but still Obama outpolled support for Kaine, although Warner did best him. The I-4 corridor was also a shocker. That region tilts a lot more center-right than it's neutral reputation would suggest and Obama's margins were surprising. At one point, I thought he was going to win in Montana, which is a state no candidate classified as a "liberal" should have any business winning any more than a candidate in New York running as a "conservative" should. And while people were calling this a paradigm shift, just two years later, everything seems to be going back to the way it was pre-2008 given the 2010 projections for particular regions.

ok, -10 then.

[KevinP]

I think that this curve is -11 or -12. -10 in the worst case but I'd wager money not lower. I think/hope that the December LSAT will produce a -12/-13 curve as it tends to have harsher curves than June/October.

[dcman06]

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.

The 160 cutoffs seem to be pretty similar to the 170 cutoffs.

2005 TESTS
PT 46/June 2005: -22 or 77 raw score
PT 47/October 2005: -23 or 77 raw
(2005 was not a curve/equating friendly year, the December '05/PT 48 LSAT had the -22/79 right mark for a 160; the 170 cutoff was -8 or 93. The total amount of LSAT's dropped by about 5% overall, so it's possible that a higher than normal proportion of these guys were well prepared for the LSAT, which may partly explain why people had to score more points for a lower score, though that doesn't say the whole story. The economy was doing pretty well, much of it due to real estate as home values were beginning to really escalate. Obviously we know to an extent why a few years later, but the foreclosure crisis didn't quite smack us just yet.

I was a college senior this year, and I remembered that some friends of mine who took the LSAT that year and went to law school for Fall 2006 were pretty antsy about the curve. Not only did these guys get lower scores on their LSATS for the same raw score, but they faced a very shitty economy when they graduated last year!)

2006 TESTS
PT 49/June 2006: -24/76 raw
PT 50/Sept 2006: -23/77 raw
(2006 wasn't much friendlier either for 160 hopefuls, and their job prospects also weren't particularly kind this summer after they graduated. But in 2006, the economy was still doing fine.)

2007 TESTS
The Free test/1st test with comparative reading/June 2007: -22/78 raw (-8 for a 170 on this test too)
PT 52/Sept 2007: -24/75 raw
2007 started scary for 160 hopefuls, but it improved a bit for the September test. The foreclosure crap began happening this time, and oil prices (along with gas) began to start creeping up like the damn Energizer bunny

2008 TESTS
PT 54/June 2008: -23/78 raw
PT 55/Oct 2008: -23/77 raw
2008 was again a pretty rough year for 160 hopefuls. Generally speaking from 2005 to 2008, 160 hopefuls had to get about 2-3 more questions right on average to get the same score as the test takers from 2000-2002. These were the last guys overall to make a well informed decision to take the LSAT right before the stock market had unstoppable diahrrea, gas was over $4/gallon in July and then dropped to below $2/gallon by Christmas in many locations, and the economy just was really bad, period though the unemployment lines also began to get larger since the national rate was about 5% in 1/2008 and over 7% by New Years Eve

2009 TESTS
PT 57/June 2009: -26/75 raw
PT 58/Sept 2009: -26/75 raw
The economy sucked and unemployment skyrocketed to over 10%. Dow Jones index dropped to the 6600's mark at one point though a lot of this was recovered by the end of the year. More people rushed to take the LSAT as their economic stimulus/bail out plan. Partly due to this, 160's hopefuls finally were able to miss on average 6 per section in order to get that score for the first time since the 2004 tests.

2010 TESTS
PT 60/June 2010: -27/72 raw
PT 61/Oct 2010: (God only knows)

The economy still sucks, unemployment is still high though it has subsided from 10.6% in Jan. to now 9.2%, and the the Tea Party movement and the Republican part are about to wreak some havoc in Congress the week after the October test results come out. The Dow Jones is now close to 11,000 again. Oil is rising again....... So the economy overall is probably improving some. But once again, the 160 hopefuls were able to average over 6 wrong a section, actually close to 7 wrong a section in June. Can this continue for October? That remains to be seen.

But if the economy improves into 2011, the good news is that there will be more jobs in many areas and if you're a Democrat, Obama is more likely to keep his job. The bad news is that the LSAT equating for 160 and 170 will probably creep up toward the 2005-2008 levels again since fewer people will take the test, and if you're a Republican, Obama is more likely to keep his job. For the superstitious, the other piece of bad news is that the 2011 test takers would mostly go to law school for Fall 2012, where they will finish their first semester and then the world will be obliterated right before Christmas

[NJcollegestudent]

I want you to be on my research team.

No you don't. As much as I love numbers crunching, I had my a$$ handed to me trying to predict 2008 election returns. It's not that I had the Presidential contest in doubt, but the margins and areas where the totals came in that bothered me. To this day, Penn, Ohio, FL, and VA still bother me. It's not that all these states that went blue was a surprise, but the locations and margins from where Obama got his support. Data I had said that in order to win all four, he'd have to poll support higher than what most major liberals are able to garner. If you look at those four states, the most successful Democratic candidates usually are more to the center/right at the state/national level. It has been this way given the make-up of those state's populaces. The "conservative" or "centrist" Democrat was always able to draw the largest share (90%+) of party supporters and the best chance of drawing a majority of moderates, while the more liberal Democrat usually was about 7-10% behind in own party support (80%).

Most surprising was OH. Obama polled better in party support and in rural "blue" areas than Ted Strickland! Not far behind was PA where Obama bested the popular moderate Rendell, taking sizable majorities in Berks and Bucks Counties, and increasing his margins in Southwestern PA despite McCain's gains on the OH/PA border. Loudon County in VA was more acceptable, but still Obama outpolled support for Kaine, although Warner did best him. The I-4 corridor was also a shocker. That region tilts a lot more center-right than it's neutral reputation would suggest and Obama's margins were surprising. At one point, I thought he was going to win in Montana, which is a state no candidate classified as a "liberal" should have any business winning any more than a candidate in New York running as a "conservative" should. And while people were calling this a paradigm shift, just two years later, everything seems to be going back to the way it was pre-2008 given the 2010 projections for particular regions.

Almost everyone underperformed in their expectations about 2008 and who was to win and by what margin. The fact that the republican was ineffective at their message, compiled with the enormous voter registration increase, which started many years before, just added on to their victory.

About 2010, i would hold off on any predictions, based on the sheer amount of doubt that is being cast on the polling methodology and the manner in which they are conducted. Honestly, the polls and pundits are confused, and i do expect minimal losses from the democrats.

[bartleby]

-12 please.

[cofc2008]

-12 please.

I agree with the scrivener.

[risktaker]

I predict -13 and I have taken Sept 09 through June 10 sadly. I thought this RC was more tricky than June RC.

[niederbomb]

About 2010, i would hold off on any predictions, based on the sheer amount of doubt that is being cast on the polling methodology and the manner in which they are conducted. Honestly, the polls and pundits are confused, and i do expect minimal losses from the democrats.

If it's anything like 2008, including the Democratic primary where HC consistently trounced Obama from March forward after her campaign got its game together (albeit too late), then the polling firms should under predict the performance of the "strong" party going into the election.

My theory: it's because more and more folks nowadays use cell phones only and ignore numbers they don't recognize, so the polling companies have more trouble getting an accurate cross-section of the people who actually vote--even the ones that do have the technology to poll mobile-only users. Personally, even if I do answer the phone and find out it's someone trying to advertise something/collect information, I immediately hang up. Anecdotal evidence only, of course. But I'm really surprised the polling companies are even as successful as they are.

Also, the enthusiasm gap should count for something.

So, I expect most of the close races to break for the Republicans, giving them 49-50 Senate seats and about 240 in the House. Alright, quote me in two weeks when it's over.

Here's a way to answer the curve question once and for all (before finding out the actual curve): For those of you who took PT 56 and 58, and this test, which one seemed harder?

56 and 58 had no one section that stood out as difficult, just like people's descriptions of PT 61 (and my overseas test), yet both are examples of modern tests that had -11 curves.

So, if PT 61 was a little bit harder, then that would indicate a -12 curve. If it was easier/about the same, then that would indicate a -11 curve.

On PT 60, I personally thought that RC and LR were rather easy, and LG had three ridiculously easy games and one ridiculously hard game. (At least for me, mulch was a gimme and I struggle with games more than most people on TLS). However, how many test takers even get to the last game? Probably a minority, a significant one most probably. So PT 60 was an easy test (for most people) with a -12 curve. It is possible October could also be a (relatively easy) test with a -12 curve. Then maybe December will be -13, and next year the curve will start declining again as the economy improves.

[motiontodismiss]

But if the economy improves into 2011, the good news is that there will be more jobs in many areas and if you're a Democrat, Obama is more likely to keep his job. The bad news is that the LSAT equating for 160 and 170 will probably creep up toward the 2005-2008 levels again since fewer people will take the test, and if you're a Republican, Obama is more likely to keep his job. For the superstitious, the other piece of bad news is that the 2011 test takers would mostly go to law school for Fall 2012, where they will finish their first semester and then the world will be obliterated right before Christmas

Double Dip FTW!!!!!

[niederbomb]

Double Dip FTW!!!!!

How about a 20-year, Japanese-style, multiple dip?

I just read the New York Times article about this and was almost too scared to invest myself in the U.S. economy by going to law school. The feeling lasted about 24 hours. Now I'm working on my apps.

[motiontodismiss]

Double Dip FTW!!!!!

How about a 2-year, Japanese-style, multiple dip?

That works too. The economy does need to recover in time for OCI, at least.