Easiest Test of All Time

[Trevor]

Games tend to be all-or-nothing.  You may make a stupid error on one question by mis-reading or something, but I've always found that if I answer them all I get them all right.  There's just aren't any judgement calls; either you see the deductions or you guess randomly.  I was freaked out after taking a June 99 practice test where I could only finish three games, but I finished yesterday with minutes to spare on both games sections.  Definitely seemed easier than any practice section I can remember, but that may just be a pressure edge.  If I manage it right, high stakes actually help me focus, so it could be that the test is really no easier but that I just had my brain in the right place and it felt that way.

[superiorlobe]

Yikes.  176 180 180 on timed preptests is pretty impressive.  Were those the only practice tests you took?  Or were those just your three best scores?

[Trevor]

Those are the first three from "10 More Actual, Offical LSAT PrepTests," with me taking them on the dining room table with a stopwatch.  I think I wound up taking six tests in all, but I stopped posting here for a while, since obsessing about things usually hurts my performance.  After reviewing my first test I figured out a few things about RC and LR (namely: make no assumptions), and that kept me in the 178-180 range until I took that June '99 test last week. I re-tooled my games strategy (less focus on constraint satisfaction and deductions, more on brute force) and did a few more games sections, but no more full tests until the real thing yesterday.

As for the scores, I don't know what to say.  There were some "nature vs. nurture" threads a while back; I've always been good at multiple-choice pressure tests.  And I've taken more formal logic that you can shake a stick at, which helps.

[superiorlobe]

What does your gut tell you that you scored on yesterday's test?

[Trevor]

There wasn't a single question that I wasn't confident I had the right answer to.  So my 'gut' says I didn't miss any.  Judging by past peformance, that means I probably missed 2-4.  This seems to be pretty constant, no matter what the curve of the test is.

But yeah, I nailed it.  Best one I ever took.

[Skittles]

i doubt that the majority of people thought that it was easy and since there is really no consensus on what was easy or difficult, then perhaps the curve will be more generous than you think.  ahhh, i hope so. 

I agree. I think it's kinda silly to worry about if the test was particularly hard or easy.  Why do you people think LSAC has experimental sections?  The designers of the LSAT go to great lengths to ensure that the test is consistant. 

And the "curve" as some people call it, is almost always the same.  I've actual taken every single test they've released and the scale is remarkably consistant, with a perfect score being a raw score of 98-100 or 99-100, then for every raw score point below this, you lose one score point until you get close to 170.  Then the test score tends to span two raw score points. 

Even though I finished the games section (though I NEVER finish a game section) I don't think it was any easier.  I think I finished because of the "test day buzz."  On the real thing your more alert and more concentrated on the test because this is the REAL THING.

The OP is just a tool who wants to make people feel dumb. 

[Skittles]

But the june test was concerned hard and the curve (which was pretty generous) reflected this ...

if the oct. test was as easy as everyone says it was, then expect a much worse curve. 

How do you figure?  I've never seen the "curve" vary by more than 2 raw points.

[zxcvbnm]

Those are the first three from "10 More Actual, Offical LSAT PrepTests," with me taking them on the dining room table with a stopwatch.  I think I wound up taking six tests in all, but I stopped posting here for a while, since obsessing about things usually hurts my performance.  After reviewing my first test I figured out a few things about RC and LR (namely: make no assumptions), and that kept me in the 178-180 range until I took that June '99 test last week. I re-tooled my games strategy (less focus on constraint satisfaction and deductions, more on brute force) and did a few more games sections, but no more full tests until the real thing yesterday.

As for the scores, I don't know what to say.  There were some "nature vs. nurture" threads a while back; I've always been good at multiple-choice pressure tests.  And I've taken more formal logic that you can shake a stick at, which helps.

I'd be interested to hear more about your games strategy. I'm about in the same boat as you are -- consistently hitting 176 - 180 on practice tests, with games always being the determining factor. I also discovered I was wasting too much time and energy making deductions up front, especially when a lot of the questions could be answered pretty simply just by eliminating the wrong choices. Do you do any up front work, or just go straight to the questions, making the deductions as needed?

[hocuspocus]

Those are the first three from "10 More Actual, Offical LSAT PrepTests," with me taking them on the dining room table with a stopwatch.  I think I wound up taking six tests in all, but I stopped posting here for a while, since obsessing about things usually hurts my performance.  After reviewing my first test I figured out a few things about RC and LR (namely: make no assumptions), and that kept me in the 178-180 range until I took that June '99 test last week. I re-tooled my games strategy (less focus on constraint satisfaction and deductions, more on brute force) and did a few more games sections, but no more full tests until the real thing yesterday.

As for the scores, I don't know what to say.  There were some "nature vs. nurture" threads a while back; I've always been good at multiple-choice pressure tests.  And I've taken more formal logic that you can shake a stick at, which helps.

I'd be interested to hear more about your games strategy. I'm about in the same boat as you are -- consistently hitting 176 - 180 on practice tests, with games always being the determining factor. I also discovered I was wasting too much time and energy making deductions up front, especially when a lot of the questions could be answered pretty simply just by eliminating the wrong choices. Do you do any up front work, or just go straight to the questions, making the deductions as needed?

Tools.

[Trevor]

The first games question in a setction can always be answered by elimination.  In general each wrong answer choice will break 1 rule, so I just go down through the rules in order and eliminate the answer choice the breaks them.

Then I draw a little diagram, and write down any deductions that leap to mind, but I've found it's better for to leap right into the questions and not spend any time pondering the setup.  Usually the first few questions I can get off the top of my head, but once the interactions start getting hairy I do best if I juct buckle down and start testing the answer choices one after the other.

On each question, there's usually a few answers that can be tossed right away to thin down your search space a little, but when it gets down to it, questions like "which of the following MUST be true" or "All of the following MAY be true EXCEPT" I just have to try everything.  Having a good diagram and a few simple deductions can make this process faster, too, so it's not like that effort is necessarily wasted, and sometimes you work out extra dependencies in the process of checking answers, which will make later questions easier and faster.  It's ugly, but it makes me go faster, at least.

Sometimes you get an intuition (turns out it's a pretty good one) about which answers to check first, too, but I can't explain how that works.

That's my strategy anyway.  Yesterday, the dogs matching question just clicked - I just saw all the dependencies after the second question - and I barely had to do any brute force work on that question, letting me pick answers directly.  But you can't plan for that sort of thing, and I like to think my method is robust enough to get them all even without a 'breakthrough.'