December 2010 Curve Prediction Poll Forum

[2011Law]

my guess is on 13/4.

[confusedlawyer]

Whoever voted -12 doesn't make any sense, December was at least twice as hard as October, and December usually has better curves. With those 2 facts I believe it is likely better, -13, -14 if we're lucky.

[2014]

Whoever voted -12 doesn't make any sense, December was at least twice as hard as October, and December usually has better curves. With those 2 facts I believe it is likely better, -13, -14 if we're lucky.

December is historically 11.5ish according to LSAT blog. For someone who thinks the test is harder than normal, 12 is a reasonable prediction.

A hard games section is memorable to test takers, but if the scores for RC, LR, LR are all normal, then the test as a whole might not be that far off the average.

I'm guessing -13 though personally.

[2011Law]

bump, since i got nothin better to do

[bartleby]

12/13. I'd be sad/happy but not surprised if it was -11/-14.

[Casey2889]

don't forget that this was a 102 question test, though, so -12 wouldn't be as generous as you think.

a -13 would put this test on par with being a little harder than october; a -14 would be super friendly; -15 is just silly.

that said, i'd take it. i crushed everything but the games, in which i nailed the first, muscled through the second (maybe missing 1-2), and then hurried through last two, blind guessing on 5-6. was PTing at 172, and the test felt better than usual (aka 174/175) until the damn games. i think i may have -10 that section alone, hoping for no worse than -13/-14 overall and/or a generous curve to crack 170.

[robotclubmember]

RC was easier than average, and both LR's were a little easier than average. That's a fact. The LG was memorably difficult. But 25% of the test being memorably difficult only just compensates for the other 75% being discernibly easier than normal.

So what. Everyone on this board can ace LG like that because it's learnable, and then one test came along that gave the majority of us a run for our money. Just because a lot of us got trounced on our strong suit (which, it isn't really a strong suit if it's everyone else's strong suit) doesn't mean that the wishful thinking is justified. It was only 25% of the test. This is a -11 curve. You're all stoned or desperate. Sorry dudes.

[2011Law]

don't forget that this was a 102 question test, though, so -12 wouldn't be as generous as you think.

a -13 would put this test on par with being a little harder than october; a -14 would be super friendly; -15 is just silly.

that said, i'd take it. i crushed everything but the games, in which i nailed the first, muscled through the second (maybe missing 1-2), and then hurried through last two, blind guessing on 5-6. was PTing at 172, and the test felt better than usual (aka 174/175) until the damn games. i think i may have -10 that section alone, hoping for no worse than -13/-14 overall and/or a generous curve to crack 170.

ouch on the 5-6 blind guesses, I had to do one, but there were a bunch that I was only like 60% sure about, so I'm guesstimating I got -5. I feel lucky actually, I had the RC LR LG LG LR order and I had to blind guess on 6 for the experimental, and was so happy that the first thing after the break was LG. I think the experimental LG was a great warm up, very happy about my ordering.

[xmrmckenziex]

RC was easier than average, and both LR's were a little easier than average. That's a fact. The LG was memorably difficult. But 25% of the test being memorably difficult only just compensates for the other 75% being discernibly easier than normal.

So what. Everyone on this board can ace LG like that because it's learnable, and then one test came along that gave the majority of us a run for our money. Just because a lot of us got trounced on our strong suit (which, it isn't really a strong suit if it's everyone else's strong suit) doesn't mean that the wishful thinking is justified. It was only 25% of the test. This is a -11 curve. You're all stoned or desperate. Sorry dudes.

That being said, it is presumed that the LSAT creates its curve based off of two large points 1. How difficult they found the test they created to be and 2. Considerations of the mass of people taking the test. While we can say that WE found it easier, we are falling into a huge sample flaw. We need to stop and think about the recent trend in the LSAT. Huge number increases in test takers are correlating with a trend towards slightly easier curves in the last two or three years. I'm not going to say one has caused the other, but I don't think it is a coincidence. More people taking the test, with a smaller percentage of them being as prepared as the kids who were originally set on law school.

I don't think -15 or anything like that is reasonable; However, I think any prediction BELOW the curve set for Oct. 2010 is as unlikely. i.e. anything below -12.

[LauraZofia]

RC was easier than average, and both LR's were a little easier than average. That's a fact. The LG was memorably difficult. But 25% of the test being memorably difficult only just compensates for the other 75% being discernibly easier than normal.

Unfortunately, I agree.

Took both Oct and Dec and while the LG difference is obvious, I think many people are overlooking the other sections' lower difficulty.

-11/-12 at the most...

[confusedlawyer]

RC was easier than average, and both LR's were a little easier than average. That's a fact. The LG was memorably difficult. But 25% of the test being memorably difficult only just compensates for the other 75% being discernibly easier than normal.

So what. Everyone on this board can ace LG like that because it's learnable, and then one test came along that gave the majority of us a run for our money. Just because a lot of us got trounced on our strong suit (which, it isn't really a strong suit if it's everyone else's strong suit) doesn't mean that the wishful thinking is justified. It was only 25% of the test. This is a -11 curve. You're all stoned or desperate. Sorry dudes.

I will keep this message, if it's -11 I'll pay your way through law school. (If I'm wrong good luck trying to find me)

[confusedlawyer]

RC was easier than average, and both LR's were a little easier than average. That's a fact. The LG was memorably difficult. But 25% of the test being memorably difficult only just compensates for the other 75% being discernibly easier than normal.

Unfortunately, I agree.

Took both Oct and Dec and while the LG difference is obvious, I think many people are overlooking the other sections' lower difficulty.

-11/-12 at the most...

RC was easier I agree, and this is coming from someone who's weakest section by far was RC, but I'm telling you LR was as such that the tricks would mess a lot of people up.

[robotclubmember]

RC was easier than average, and both LR's were a little easier than average. That's a fact. The LG was memorably difficult. But 25% of the test being memorably difficult only just compensates for the other 75% being discernibly easier than normal.

So what. Everyone on this board can ace LG like that because it's learnable, and then one test came along that gave the majority of us a run for our money. Just because a lot of us got trounced on our strong suit (which, it isn't really a strong suit if it's everyone else's strong suit) doesn't mean that the wishful thinking is justified. It was only 25% of the test. This is a -11 curve. You're all stoned or desperate. Sorry dudes.

That being said, it is presumed that the LSAT creates its curve based off of two large points 1. How difficult they found the test they created to be and 2. Considerations of the mass of people taking the test. While we can say that WE found it easier, we are falling into a huge sample flaw. We need to stop and think about the recent trend in the LSAT. Huge number increases in test takers are correlating with a trend towards slightly easier curves in the last two or three years. I'm not going to say one has caused the other, but I don't think it is a coincidence. More people taking the test, with a smaller percentage of them being as prepared as the kids who were originally set on law school.

I don't think -15 or anything like that is reasonable; However, I think any prediction BELOW the curve set for Oct. 2010 is as unlikely. i.e. anything below -12.

There is no consideration of the mass of people taking the test on test day. The psychometricians work that out through the experimental sections that they analyze. The curve is predetermined, and usually it conforms fairly accurately to what the percentiles are. Those guys know what they are doing. However, there are times in which a 170 may be 97th percentile, and other times when it may be 98th percentile, so it doesn't always work out the same. If they curved it after the fact, it would be the same though, the fact that it isn't always just shows that no, they don't care about the mass of people, as you call it, taking it on test day. At least the percentile you are in for each administration is reported to law schools along with your score by LSAC (though I don't think they care about that).

And there is no statistical evidence of the curve getting easier in the last few years. In fact, the opposite is true:



-15 is completely unreasonable. These posts are very frustrating because I just took the test too, and I'm on pins and needles like everyone else. That said, I don't like the idea of people getting my hopes up, or everyone else's hopes up, with these bogus predictions, because the facts do not bear them out.

It is very possible that the LR question that the most people got wrong will be excluded from scoring bringing it down to 101, which helps. That plus a -11 curve is all the breaks you guys are going to get. Deal with it. 75% of the test was just plain easy. One hard section alone has never resulted in a curve as generous as you predict, unless all the other sections were average difficulty. But they weren't.

[2014]

And there is no statistical evidence of the curve getting easier in the last few years. In fact, the opposite is true:

Okay I don't know where you are from, but where I come from "last few years" does not encompass back to 1991...
I would argue that it is perfectly reasonable to state that over the last few years the curve is getting more friendly, as shown by the blue line I have drawn over your graph. Now I didn't run the actual regression or whatever, but I think I'm probably close, perhaps a little too steep.

--ImageRemoved--

[xmrmckenziex]

And there is no statistical evidence of the curve getting easier in the last few years. In fact, the opposite is true:

Okay I don't know where you are from, but where I come from "last few years" does not encompass back to 1991...
I would argue that it is perfectly reasonable to state that over the last few years the curve is getting more friendly, as shown by the blue line I have drawn over your graph. Now I didn't run the actual regression or whatever, but I think I'm probably close, perhaps a little too steep.

--ImageRemoved--

My sentiments exactly. I understand where you are coming from and it is a completely valid argument. But you can't say that there hasn't been a recent trend in the curve getting just a little more helpful. We can argue how recent, but for some length of time it has.

[pinkcamellia]

I voted for -12 but I'm hoping for -13. The LG section was very difficult yesterday, as most people are saying, but I found the other sections to be of average difficulty. I'm sure we'll all do just fine!

[robotclubmember]

And there is no statistical evidence of the curve getting easier in the last few years. In fact, the opposite is true:

Okay I don't know where you are from, but where I come from "last few years" does not encompass back to 1991...
I would argue that it is perfectly reasonable to state that over the last few years the curve is getting more friendly, as shown by the blue line I have drawn over your graph. Now I didn't run the actual regression or whatever, but I think I'm probably close, perhaps a little too steep.

--ImageRemoved--

Yeah, I can tell you didn't run the regression by looking at that. Good regression analysis usually accounts for the high and low outliers, which you do not do, and your methodology involves starting from the lowest point, which visually is comforting but maybe not the most accurate. Notice how the troughs go way farther beneath your regression line than the peaks do. What does that tell you about it?

That said, I can see your point that over the last three years, it might be slightly trending up. But your regression line is as misleadingly optimistic as the -15 predictions.

I'm still sold on -11.

1 year trend, 3 year trend, 10 year trend, who cares. The test as a whole just wasn't more difficult than average. It just had one hard section.

[2014]

Yeah I realized after I posted that starting at an extreme was incorrect but couldn't be bothered to open up paint again . Realistically it would be flatter though still upward trending.

I stand by -12/-13 regardless, but recognize that my optimism might cloud things. One thing for sure is that I would be SHOCKED if it were -10 or worse.

[robotclubmember]

Yeah I realized after I posted that starting at an extreme was incorrect but couldn't be bothered to open up paint again . Realistically it would be flatter though still upward trending.

I stand by -12/-13 regardless, but recognize that my optimism might cloud things. One thing for sure is that I would be SHOCKED if it were -10 or worse.

Well hey, I won't be complaining if you're right!

[brobrah]

While I thought that my performance on the other three sections will match my practice average, I would say as a whole the LR sections are starting to get harder. My 55+ PT scores roughly stayed the same as the tests in the 40s and early 50s, but I believe my actual skill level went up during the course of those tests, and the tests just got progressively harder to match my increased ability. The LR is just having more funky question types with vague answer choices. They are really trying to get rid of the "good" answers and just make the correct answers the ones that are "less wrong" than the others. For example, strengthen and weaken questions have gotten a lot harder because the right answers only barely weaken or strengthen - making them really hard to discern from the other wrong choices. The necessary assumption stims are not making the argument gaps very obvious anymore, and answer choices all look very good, and you pretty much need to use the negation technique to figure out which assumption is really the "necessary" one. Basically every question has a lot of trap answer choices and you really need to know your shit to figure out which ones are correct. Compared to the types of questions I remember from the bibles, there is definitely a huge difficulty gap between those books and some of the current stuff. If I hadn't run the PT45-61 marathon I would have had a huge shock at the increase in difficulty that the 55+ tests have.

So I agree the trend on the curve is definitely upwards in recent years. RC (my worst section usually) I thought was a little more forgiving, or perhaps I was just really able to engage with the passages this time. The RC definitely didn't make up for the LG surprise, and LR was on par with recent tests, so I doubt we will be seeing an average curve. Probably -13.

[pattonthicke]

Everyone keeps saying the RC was easy but i found it to be difficult. Considerably more difficult than oct or june, though comparable to many sections in the 50s. That said it could just be that i had a problem with this section and everyone else thought it to be easy. Just bad luck for me i guess. LR was more of the same and i already suck at games so it felt the same to me. Im still thinking -13 at least b/c it just seemed more difficult than pt 60 or 61.

[xmrmckenziex]

Yeah I realized after I posted that starting at an extreme was incorrect but couldn't be bothered to open up paint again . Realistically it would be flatter though still upward trending.

I stand by -12/-13 regardless, but recognize that my optimism might cloud things. One thing for sure is that I would be SHOCKED if it were -10 or worse.

Well hey, I won't be complaining if you're right!

At this point, none of us can really be sure. I mean, the LSAT seems to consistently do weird things without rhyme or reason. All we can do now is really wait and see.

[brobrah]

I think what I meant by RC being more forgiving was that the passages were a bit more straightforward to comprehend, but the questions were still just as hard. I think only the 1850s one I had a bit of problem reading but the questions were a bit easier to make up for it.

[veragood]

I'm expecting/hoping for a -13 curve since I believe in positive thinking : )

But.. if I had to put money on it, I would say -12.

I would put the odds of -14 near 3%, obviously I would be insanely happy if it ended up being that.

[DrackedaryMaster]

The Oct 2010 test was only the 3rd time ever since 1991 that the raw score required to get 170+ required more than 1 correct answer from the June test. And this year's occurrence may have had something to do with the fact that June '10 had only 99 questions.

Looking at the raw score comparisons of the Oct/Dec tests, it usually goes down in December requiring fewer correct answers to hit 170, but there have also been some years in which there were wild swing upward and downward by as much as 5 answers! The fact you guys have said this test has 102 questions seems to lead to possibly a -13/-14 type scale.