December 2010 Curve Prediction Poll Forum

[FlanSolo]

That question was very odd (as was the chef/virus one), but not particularly challenging. If you use basic logic, it's fairly simple to deduce the correct response. Ditto for the chef/virus question.

Yeah, I'm not sure we're disagreeing about anything. There's really no basis to expect that any questions are going to get removed. Questions get removed when the "wrong" scorers are getting a question right, not when a question deviates from traditional LSAT norms. My sense is that there are more than enough high scorers who managed to avoid being brainwashed by their test prep to make up for the small minority of people who score high and got such "odd" questions wrong because they over thought it.

I'd hate to say it, but my guess is that this will hold true for the LG from this test as well. That's not to suggest they weren't difficult questions, just that smart people with adaptable strategies probably fared better than smart people with rigid strategies. This site probably tends to attract a disproportionately higher amount of smart people with rigid strategies and hence there's more of a sense that these games were a bloodbath. Again, that's not to say the LG weren't difficult or that they weren't the hardest section, just that it's probably a bit less of a massacre than people are expecting

All of that said, I'm sticking with my guess of -14, owing to it being a more difficult and longer test than October. For the reasons I just stated though, I wouldn't count on the curve being any easier than that.

[kkklick]

The more I remember the test, the more I see the curve at -14. Just way too many difficulty variables that would cripple the average test taker and hurt above average scorers.

[Sandro]

The more I remember the test, the more I see the curve at -14. Just way too many difficulty variables that would cripple the average test taker and hurt above average scorers.

Do you think LSAC determines difficulty by something like % of takers who miss questions? The more I think about it some of the RC and definitely LG would have just torn up most people. And when I say most people, im talking the 80ish percent of test takers that TLS really doesnt represent. Some people have been talking as if LSAC looks at who misses the questions to determine their difficulty but I think that is bogus and wouldn't work to well.

[2014]

The more I remember the test, the more I see the curve at -14. Just way too many difficulty variables that would cripple the average test taker and hurt above average scorers.

Do you think LSAC determines difficulty by something like % of takers who miss questions? The more I think about it some of the RC and definitely LG would have just torn up most people. And when I say most people, im talking the 80ish percent of test takers that TLS really doesnt represent. Some people have been talking as if LSAC looks at who misses the questions to determine their difficulty but I think that is bogus and wouldn't work to well.

There was a good explanation of it on LSAT blog. Say on average 95% of 170+ scorers get a question right but on one question, 20% of the 170+ scorers missed it as an experimental, that question will be considered a difficult question then and will help the "curve" for 170 by some marginal amount only LSAC knows. If 57% of 155 scorers get a question right, and on the experimental section 60% of 155 scorers get it right, then it is possible that the "curve" for 155 will be worse for them because for 155 scorers, the question was actually easier than normal.

So obviously one question isn't going to be big enough to shift it from -11 to -12 or w/e, but I think that's how they equate it, adjusting slightly for variations in any given question, resulting in an overall adjustment either + or - from their baseline. In this case we are assuming that the average 170 scorer did worse on these sections as experimentals than they should have on average

[kkklick]

The more I remember the test, the more I see the curve at -14. Just way too many difficulty variables that would cripple the average test taker and hurt above average scorers.

Do you think LSAC determines difficulty by something like % of takers who miss questions? The more I think about it some of the RC and definitely LG would have just torn up most people. And when I say most people, im talking the 80ish percent of test takers that TLS really doesnt represent. Some people have been talking as if LSAC looks at who misses the questions to determine their difficulty but I think that is bogus and wouldn't work to well.

No point in speculating LSAC's equating philosophies we'll never know. I've just felt that every curve fits most tests well with a few exceptions. This felt like a -14.

[Sandro]

--LinkRemoved--


It might be a little wishful thinking but even if I bombed LG again with another -8 and kept my RC at -3 like in October, all it would take is 2 less stupid mistakes on LR for a ... 167 ! No -14 curved test that I can see has -18 at 101 less than a 167.

I would cry if this happens

[fosterp]

The more I remember the test, the more I see the curve at -14. Just way too many difficulty variables that would cripple the average test taker and hurt above average scorers.

Do you think LSAC determines difficulty by something like % of takers who miss questions? The more I think about it some of the RC and definitely LG would have just torn up most people. And when I say most people, im talking the 80ish percent of test takers that TLS really doesnt represent. Some people have been talking as if LSAC looks at who misses the questions to determine their difficulty but I think that is bogus and wouldn't work to well.

That is exactly what they do.

[well-hello-there]

For realz.

I am not even so sure that PT57 was more difficult than PT62. PT57 was a slam dunk -0 for me. PT62, had to blind guess on two because of time constraints (also got a splitting headache right before PT62 LG... thank god LG was section 5 for me).

PT62 as a whole is easily in the top five hardest games sections. It's right up there in the league of new and used CD's and mauve dinosaurs.

The dinosaur game IS ranked in the top 5 hardest games and the curve on that 101 question test is a -11. In my personal experience, this was an easy LG section. For ME, it wasn't until the 3rd time I did PT 57 LG section that I was able to finish it in under 35 minutes. Everybody has their own strengths but for you to have a "slam dunk -0" on PT 57 LG and then say that this LG was hard doesn't add up. I don't doubt that you did bad but I think it had to be all psychological. -0 the first time on PT57 LG means that you are a LG badass. There's no getting lucky on that one the first time. Without the stress of test day, you will see that these games were easy.

[2014]

If you figured out the Mauve Dinosaur limitation though, that game was doable in well under 8:45 if it all clicked for you.

The games on LG for even the average high scorer were time drainers. The ones who finished were the ones with good intuition in deciding where to start their hypos, which is indeed a good skill to have, but isn't usually as necessary as it was Saturday.

[aesis]

If you figured out the Mauve Dinosaur limitation though, that game was doable in well under 8:45 if it all clicked for you.

The games on LG for even the average high scorer were time drainers. The ones who finished were the ones with good intuition in deciding where to start their hypos, which is indeed a good skill to have, but isn't usually as necessary as it was Saturday.

This. If you weren't already well-attuned to using hypos as a initial jump start to your games - as opposed to going on inference hunts -- then you lost time. A lot of time. And hypo-crazy games make you lose even more time. Ugh, what a shitty LG section.

[skip james]

I have done every single prep-test available and the LG wasn't even in the top 25 hardest sections. And I'm not that good on LG. (-4 in October) RC was stupid easy compared to October. LR, really not that bad either. Altogether, it reminded me of PT's 1-20 in easiness and this is definitely the easiest one since September 2009. My lowest PT score was 161 for PT-55 taken on September 09, 2010 and my highest PT score was 175 for the June 2007 test which I took only 5 days ago. I think the curve will be a -12 and that I will get a 170.

I don't agree. I've taken the test several times, scored in the 99th twice and I found this test fairly difficult.

RC was tricky, LR was pretty typical, the first section being more difficult than the second. Games were pretty standard in my opinion, though the type of game that stained glass was more reminiscent of the games from the 30s, and probably screwed people who focused more or less on the 'standard' linear/grouping games.

I did notice lots of sufficient assumptions though. Trickier han normal too.

[skip james]

I bet the scale is fairly generous -13 seems reasonable, -14 wouldn't surprise me, -15 is kinda a stretch bit I don't think impossible given the fact that there were 102 questions.

You have to into account that the difficultly of the test isnt limited to the difficulty of the questions themselves but Lso the increased time constraints involved in a 102 question test.

[KevinP]

As the days go by.... I think the curve will not matter for me. A 120 on a -15 curve will still be a 120 on a -11 curve ;/

[FlanSolo]

If you figured out the Mauve Dinosaur limitation though, that game was doable in well under 8:45 if it all clicked for you.

The games on LG for even the average high scorer were time drainers. The ones who finished were the ones with good intuition in deciding where to start their hypos, which is indeed a good skill to have, but isn't usually as necessary as it was Saturday.

I agree. I posted a longer explanation in another thread, but I think people are letting the conventional wisdom about Dinos color their judgment and putting PT 57 on bit of a pedestal. People are seeming to forget that the remaining games on PT 57 weren't so rough. I forget the themes, but there was one very, very simple linear game, one pretty simple grouping game (both of which came before Dinos), and the Awards game after Dinos was pretty straightforward as well.

On December, you had one very simple game (that came last!), one that was relatively simple but very space constrained (definitely more than Awards), one pretty tricky, time intensive game (conferences), and one straight up difficult game in stained glass. This was a damn hard section.

[robotclubmember]

For realz.

I am not even so sure that PT57 was more difficult than PT62. PT57 was a slam dunk -0 for me. PT62, had to blind guess on two because of time constraints (also got a splitting headache right before PT62 LG... thank god LG was section 5 for me).

PT62 as a whole is easily in the top five hardest games sections. It's right up there in the league of new and used CD's and mauve dinosaurs.

The dinosaur game IS ranked in the top 5 hardest games and the curve on that 101 question test is a -11. In my personal experience, this was an easy LG section. For ME, it wasn't until the 3rd time I did PT 57 LG section that I was able to finish it in under 35 minutes. Everybody has their own strengths but for you to have a "slam dunk -0" on PT 57 LG and then say that this LG was hard doesn't add up. I don't doubt that you did bad but I think it had to be all psychological. -0 the first time on PT57 LG means that you are a LG badass. There's no getting lucky on that one the first time. Without the stress of test day, you will see that these games were easy.

Lol. "I don't doubt that you did bad." Thanks! Haha.

But seriously. I don't think stained glass was harder than mauve dinosaur, nor was conference. But both of them together, plus the game with no room to write, made PT62's LG overall harder than PT57's. That said, I don't think it was stress. I was highly non-stressed and it is likely I missed less in both LR's and RC combined than in LG (LG probably -4). LG was just that hard.

Anyway, it adds up to me, lol. Everyone has an on or off moment, and yes, after the test I will probably realize there was a better way to do it... much like 90% of the chaps on this forum. But it wasn't psychological. It was ostensibly, objectively more difficult. Perhaps you lucked out and everything clicked right away... just as even monkeys fall from trees, even blind squirrels find nuts.

I'm not trying to convince you, you have your opinion. But for other people browsing or lurking who are wondering if the LG was really as hard as they thought it was, trust me, it was. I can definitely say with a high level of certainty that this LG was harder than at least 90% of all that have preceded it. That said, I think the difficulty of LG was compensated for by ease of LR and RC and you all may be surprised to see yourself scoring a little better in those sections to make up for LG.

[kkklick]

I have done every single prep-test available and the LG wasn't even in the top 25 hardest sections. And I'm not that good on LG. (-4 in October) RC was stupid easy compared to October. LR, really not that bad either. Altogether, it reminded me of PT's 1-20 in easiness and this is definitely the easiest one since September 2009. My lowest PT score was 161 for PT-55 taken on September 09, 2010 and my highest PT score was 175 for the June 2007 test which I took only 5 days ago. I think the curve will be a -12 and that I will get a 170.

I don't agree. I've taken the test several times, scored in the 99th twice and I found this test fairly difficult.

RC was tricky, LR was pretty typical, the first section being more difficult than the second. Games were pretty standard in my opinion, though the type of game that stained glass was more reminiscent of the games from the 30s, and probably screwed people who focused more or less on the 'standard' linear/grouping games.

I did notice lots of sufficient assumptions though. Trickier han normal too.

I noticed this is well, assumption questions are my ace in the hole for LR, but I know for sure I got one of them wrong and had trouble on another. There were a few that didn't follow the usual assumption techniques.

In regards to LG, mauve dinosaurs killed me the first time until I realized that I overlooked that one of the toys could be any color. That seemed to open things up for me on that game. In terms of difficulty I would put them on par. Both games sections from 57 and 62 had 2 difficult ones but were difficult in different ways. 57 had a hybrid and a difficult matching game, while 62 had simple structured games but still difficult.

[androstan]

I nailed the dinosaur game the first time I took it and was suprised to read that it was considered so hard. There are only a few basic templated solutions to that game, a lot of each solution is given to you. It's more constrained than it might first appear.

On the other hand, on this set of LG, it wasn't really possible to deduce a few straightfoward templates. It was all hypos, especially for stained glass. No, the games weren't super "hard" in the sense that you had to make deep inferences or connections, but they were time-draining. For me, who has always struggled to finish the games in time (but generally had excellent accuracy), this LG was a killer. If I had another 5 minutes I would have done very well on this set of LGs.

I actually skipped game 1 because it was space constrained. By the time I got to it the proctor called 5 minutes. I think I had time to read the rules, diagram, and work the first two questions before I had to start guessing. I also had left 2 blank in a previous game (can't remember which, but it wasn't stained glass, probably conferences). I think that was, like, 6 guesses due to time. The last game I remember knocking out. Conferences I remember draining my time but still being straightfoward. Stained glass I remember REALLY draining my time and being a little less straightfoward. The first game, utilities, was actually pretty easy (a bit harder than game 4) but I ran outta time. I wish I'd done it first :/.

I think a lotta people in the low 170s and upper 160s struggle for time on logic games. It's a learnable section but a person's brain can only process so fast. I think that, when this was experimental, even some top scorers were guessing on a few due to time. I felt like the RC was slightly easier than normal and the LR's were average. The LG's are going to pull the curve down though. At least to -13 imo.

[JohnnyBlues]

First time posting on this site, although I've spent a lot of time on it for the past three months studying.

With that said, I couldn't agree more with the previous post. Exactly the same thing happened to me except I THINK I ended up only having to straight up guess on four. Never had to guess on more than 2 in LG before including the Muave Dino game (which I ended up with -0 on).

Just to add to the solidarity of this forum, I was averaging a 173 PT and scored a 178 on PT61 last Thursday only to leave this test feeling like I did worse than my diagnostic. Don't trust your worst feelings, more often than not they betray you.

Good luck to everyone...

[bigkahuna2020]

I have done every single prep-test available and the LG wasn't even in the top 25 hardest sections. And I'm not that good on LG. (-4 in October) RC was stupid easy compared to October. LR, really not that bad either. Altogether, it reminded me of PT's 1-20 in easiness and this is definitely the easiest one since September 2009. My lowest PT score was 161 for PT-55 taken on September 09, 2010 and my highest PT score was 175 for the June 2007 test which I took only 5 days ago. I think the curve will be a -12 and that I will get a 170.

I don't agree. I've taken the test several times, scored in the 99th twice and I found this test fairly difficult.

RC was tricky, LR was pretty typical, the first section being more difficult than the second. Games were pretty standard in my opinion, though the type of game that stained glass was more reminiscent of the games from the 30s, and probably screwed people who focused more or less on the 'standard' linear/grouping games.

I did notice lots of sufficient assumptions though. Trickier han normal too.

I noticed this too...not a lot lot of necessary ones....which sucks because I am shite at diagramming

[well-hello-there]

Perhaps you lucked out and everything clicked right away.

Maybe. It's just that I'm typically very susceptible to defeat by even medium-hard LG sections.

As far as assumptions are concerned: I feel like I had a breakthrough on sufficient v. necessary assumptions a few weeks before the test. It seems so obvious to me now that I can consciously identify the differences but before I figured it out, I was only average on assumption questions. This may sound dumb to a lot of you but I know it would have helped me a ton---for necessary assumption questions (eg. which one of the following is assumed by the argument.) many times the right answer choice will be the most basic of assumptions and will come off as sounding so weak and irrelevant as to be outside of scope. An example that comes to mind is a LR question where it was discussing a conclusion or a health recommendation that doctors had given based on their study of a group of patients. I think some were smokers and those study participants had self identified as smokers. The right answer was something like, "those who had identified themselves as smokers had smoked in the past" or something stupid like that but meaning that the study participants didn't lie about their smoking history. There might still be huge gaps in the author's argument but at a minimum, that would have to be assumed because if participants had never smoked, then any conclusion doctors made based on the assumption that they had smoked would be wrong. The point is that it is such an obvious assumption that I skimmed right past it and chose the wrong answer. Now when I tackle that question type, I start at the most extreme required assumption in my head like...cigarettes do in fact exist. Of course few answers are going to be that easy but thinking from that extreme starting point gets my mind ready for that type of question. Look for at least one of the wrong answers to be an answer that would be right if it were a sufficient assumption question.

For sufficient assumption questions (the conclusion follows logically if which one of the following is assumed), I start at the other extreme. The right answer will many times make the conclusion true AND THEN SOME. I think that a lot of times the right answer isn't obvious because it just sounds too extreme or outside of scope because it affects so much more than just the conclusion. But the point is that it is relevant to the conclusion. As soon as I see the question stem saying "if assumed" I get ready for answer choices that are going to seem outside of scope. Realize that you are looking for an answer that will 100% for sure definitely make the conclusion follow. To make an argument that sound, the answer will usually contain very strong/definite language. Look for strong keywords like "All" "Will" "Must" "No" "None" in the right answer and weak keywords like "some" "most" "should" "can" "might" in the wrong answers.

[kkklick]

I think the timing of your post couldn't be much worse. But all in all good points for those struggling with assumption questions. I'm very good at them but the difficulty of the LR section as a whole threw all my studying techniques out the window.

[well-hello-there]

I don't doubt that you did bad but I think it had to be all psychological.

Lol. "I don't doubt that you did bad." Thanks! Haha.

Sorry, I didn't mean anything by that. It was irrelevant to the point I was trying to make. What I should have said was, "I think your difficulty with the section should be ascribed to test day psychological factors rather than a lack of analytical reasoning abilities."