Guess the forever unknowable curve to the Feb 10 LSAT Forum

[februaryftw]

I know there is a poll for difficulty of exam, so I hope this isn't too redundant. Just wondering what others guess the curve of the exam was, especially those who have taken at least the last 5 exams in test like conditions. I realize we'll never know, but I've taken quite a few exams and usually have a sense of whether the test was more or less difficult.

A refresher of past curves:

Dec 09: -14
Sep 09: -11
Jun 09: -11
Dec 08: -11
Sep 08: -9
Jun 08: -10

My take:

I think the games seemed to trip a lot of people up, but no one game was terrible--just a lot more time consuming. I think this ensures it is at least a - 10. Logical reasoning seems to be average difficulty at worst, maybe on the easy side. Reading comprehension also seemed average. Is Sep 09 a good comparable?

I think the curve was probably around a -10 - -11.

[keg411]

People usually don't know how easy/tough LR is and that usually makes/breaks the curve. A tough games section rarely matters.

[februaryftw]

People usually don't know how easy/tough LR is and that usually makes/breaks the curve. A tough games section rarely matters.

Really? I've always thought a tough games section seems to have a decent impact on the curve. Are you mainly basing this on the fact that LR is half the test? Or?

[charlesxavier]

From what I have read on here (TLS) it seems that the curve is based on how previous test takers did on the sections that are in your test when they were the experimental section. So, I would think that means that the curve takes the difficulty of all sections into consideration.

[Pricer]

I'm guessing -10. I feel like it would have possibly been a -11 with 101 questions. But I really have no idea. Expect -7, pray for -14.

[jks289]

Does the 100 questions istead of 101 affect the curve at all? For better or worse? I voted -14 but that is because I desperately need a generous curve. Please, LSAT gods.

[februaryftw]

From what I have read on here (TLS) it seems that the curve is based on how previous test takers did on the sections that are in your test when they were the experimental section. So, I would think that means that the curve takes the difficulty of all sections into consideration.

Yes, this is true, but what is unknown (I think) is whether certain sections of the exam tend to have a disparate impact on the curve. It might be that test takers have different abilities when it comes to harder LG, LR, or RC, or that the distribution of credited responses are not equivalent across the sections. If all sections are relatively equivalent in how much they vary test to test, then obviously LR is the most important because it is half the test. But it might be, for instance, that the LG scores have more variance than do the LR scores.

[evergirl23]

I put -10, though I would love it to be higher. I just honestly didn't think the test was that difficult. Though talk to me when scores come out and I may be singing a different tune.

Also, it may not be forever unknowable - LSAT has eventually released some February tests. Just not very many.

[7ED]

.. couldn't we just figure out the curve by comparing scores vs scaled scores from different people after we get our marks back?

[februaryftw]

.. couldn't we just figure out the curve by comparing scores vs scaled scores from different people after we get our marks back?

I thought we only received the scaled score, so we won't know how many questions we got wrong.

[Pricer]

.. couldn't we just figure out the curve by comparing scores vs scaled scores from different people after we get our marks back?

I thought we only received the scaled score, so we won't know how many questions we got wrong.

I keep hearing it both ways. I'll go check LSAC for the real answer.

[imisscollege]

am I the only one who thought RC was not easy at all and LG was completely fine?

I did LG no problem with 30 seconds to spare although there was many plug and chugs at the end of the test which was a bit tough I guess

But I usually finish RC with time to spare and I could barely do that and had to do the language passage in 5 minutes.

[Pricer]

source: https://os.lsac.org/Release/help/information.aspx#LSAT

February tests receive the following:

Item Response Report (score only)
Item Response Report Information
Writing Sample

As far as what that actually means, I'm not sure.

[evergirl23]

source: https://os.lsac.org/Release/help/information.aspx#LSAT

February tests receive the following:

Item Response Report (score only)
Item Response Report Information
Writing Sample

As far as what that actually means, I'm not sure.

It means you get your raw score only - you don't get your answers or the credited answers, nor do you get the numbers you got right or wrong.

This is per the Item Response Report Additional Information PDF that I received after the December administration. It states that on your Item Response Report, those who take undisclosed administrations will have blank sections for the Question Response Data (#s right and wrong) and the LSAT Question Responses (the grid of your answers and the credited answers). You will also not receive your answer sheet, the disclosure book or the conversion table.

[somewhatwayward]

where's the option for -5?

oh, and you don't get your 'raw' score - that's the number of questions you got right; you get your 'scaled' score, which is the conversion of the raw score onto the 120-180 scale

[BigA]

confused by these numbers

[dynomite]

confused by these numbers

Yeah, I was way back when as well. But I didn't have the courage to admit it.

Unless I'm mistaken, the "curve" number is just the number of questions you can get wrong and still get a 170. It takes into account a lot of data, but basically it attempts to control for different variables and allow law schools to compare one student who took the Sept 09 test to another who, for instance, took the Feb 10 test.

So a -14 curve (the one in December) meant that people could miss 14 questions and still get a 170.

[februaryftw]

where's the option for -5?

oh, and you don't get your 'raw' score - that's the number of questions you got right; you get your 'scaled' score, which is the conversion of the raw score onto the 120-180 scale

Well the max is ten options, so I had to leave out -4 and -5. I actually put in the two extremes because I figure anyone voting for -7 or -16 is being unreasonable/being too hopeful/not taking the poll seriously, so a trimmed mean (throw out the extreme values) in this case probably better reflects those people who answered the poll more honestly.

[Pricer]

The curve won't be -5 or -6, it hasn't been that low in any recent test.

[raperez129]

Please LSAT gods, make it a 12 or more. I didn't think it was too bad, except for the games. I thought they were harder/more time consuming than most recent released tests. And I thought one of the args was a bit hard, but I'm hoping that one was experimental.

[somewhatwayward]

The curve won't be -5 or -6, it hasn't been that low in any recent test.

hi, my name is sarcasm, nice to meet you

[BigFatPanda]

I know there is a poll for difficulty of exam, so I hope this isn't too redundant. Just wondering what others guess the curve of the exam was, especially those who have taken at least the last 5 exams in test like conditions. I realize we'll never know, but I've taken quite a few exams and usually have a sense of whether the test was more or less difficult.

A refresher of past curves:

Dec 09: -14
Sep 09: -11
Jun 09: -11
Dec 08: -11
Sep 08: -9
Jun 08: -10

My take:

I think the games seemed to trip a lot of people up, but no one game was terrible--just a lot more time consuming. I think this ensures it is at least a - 10. Logical reasoning seems to be average difficulty at worst, maybe on the easy side. Reading comprehension also seemed average. Is Sep 09 a good comparable?

I think the curve was probably around a -10 - -11.

I thought the rule about curve is base on how people did on the previous LSAT. Considering September 2009 was the most taken, in term of numbers, it is reasonable to assume a lot of people took it cold and thus the mean was tanked. As a result, people in December benefited from the curve.

[bakemono]

god if it's even just -12, not that absurd -14 from that easy as hell PT 59, i'm in 170 land.

[mazz]

While I don't think it will be a -14, I don't think we'll know until we get our results. I did well on dec. 2009, except for one LR section that completely destroyed me. 11 or 12 is what I'm guessing. Could be higher though, definitely not lower IMO.

[raperez129]

This may be a silly question, but after the release, how willl we know what the curve was?