Top Law Schools

So I spent the past 6 weeks studying for the LSAT and took it yesterday, and turns out you can cancel your score if you think you did badly. Seems like a very common thing, now that I read these forums. I do not wish to cancel my score. However, I am curious as to what happens to someone's test once they do cancel it. Is it still graded and thrown into the overall bell curve without actually reporting the score back to the individual, or is it as if they never took the test and only non-canceled scores make up the entire left side of the curve? Let's face it - about half the people who take the LSAT are going to score a 150 or lower. I personally find it unfair if some of the tests from people who get a lower score than they originally anticipate just being discarded, when they are still representative of the overall population that took the test. Does anybody know the answer to this? I have searched LSAC's website and can't find one.

Thanks.

I would guess the score is not in included. Good question though.

...

paul34 wrote:Can you really count those people? It wouldn't be fair to do so, since they didn't really *do* the test, and so it wouldn't be reflective of how people *did* on the test.

Where do you think the 120's come from? Every year 12-15,000 people score in the bottom 10th percentile. Even if you just randomly guess you're likely to get abouta 125-126, so I'd imagine a large percentage of the 120-125's left large portions of their test undone.

Some good thoughts here so far, but it might be all wishful thinking. Being as how I'm now waiting for my score, I hope to high heaven that those unfinished tests are included in computing my score. For those who already have a score from a previous test, I'm sure that they would rather think that their percentile is included only agaist those who finished the test and did not cancel their scores rather than believe that their 160 may have been a 150 if...

i suspect that cancelled scores are never graded at all

Remember that it's not a curve. They determine the scoring grid in advance. So whether anyone else cancels or not makes no difference to your score. No one else's performance on your test affects your score.

As far as the equating sections go, I would guess that people who actually finish the test get scored internally by LSAC in order to give them more data on the equating sections for future test-takers. But it wouldn't actually matter either way, statistically.

Wakamusha wrote:Some good thoughts here so far, but it might be all wishful thinking. Being as how I'm now waiting for my score, I hope to high heaven that those unfinished tests are included in computing my score. For those who already have a score from a previous test, I'm sure that they would rather think that their percentile is included only agaist those who finished the test and did not cancel their scores rather than believe that their 160 may have been a 150 if...

You have a big, yet very common, misconception about how the LSAT is scored. The LSAT is NOT graded on a curve. Scaled scores achieved by people that don't cancel ARE NOT curved. How the population of people that took the same test as you performed DOES NOT determine the scoring scale. There is a big difference between a test being graded on a conversion scale and a test being graded on a curve.

The LSAT is a standardized test. 150, 160, 165, 170, etc. has to mean the same thing every time (an ability/performance level) and be comparable across administrations/test forms, no matter which test it is from and no matter whether the group of people that took that test overall were a bunch of geniuses or a bunch of dim wits.

Other people canceling has no effect on your scaled score.

Only affects you if the other person accidentally cancels your score---something that is not easily done.

Jeffort wrote:

Wakamusha wrote:Some good thoughts here so far, but it might be all wishful thinking. Being as how I'm now waiting for my score, I hope to high heaven that those unfinished tests are included in computing my score. For those who already have a score from a previous test, I'm sure that they would rather think that their percentile is included only agaist those who finished the test and did not cancel their scores rather than believe that their 160 may have been a 150 if...

You have a big, yet very common, misconception about how the LSAT is scored. The LSAT is NOT graded on a curve. Scaled scores achieved by people that don't cancel ARE NOT curved. How the population of people that took the same test as you performed DOES NOT determine the scoring scale. There is a big difference between a test being graded on a conversion scale and a test being graded on a curve.

The LSAT is a standardized test. 150, 160, 165, 170, etc. has to mean the same thing every time (an ability/performance level) and be comparable across administrations/test forms, no matter which test it is from and no matter whether the group of people that took that test overall were a bunch of geniuses or a bunch of dim wits.

Other people canceling has no effect on your scaled score.

So how does the LSAC make a scale in advance? Does it come from non-disclosed Feb test? I understand it is not curved but I do not know why certain exams are 10 for 170, others 12 for 170. Is it just the matter of difficulty assumed by the LSAC?

lovejopd wrote:

Jeffort wrote:

Wakamusha wrote:Some good thoughts here so far, but it might be all wishful thinking. Being as how I'm now waiting for my score, I hope to high heaven that those unfinished tests are included in computing my score. For those who already have a score from a previous test, I'm sure that they would rather think that their percentile is included only agaist those who finished the test and did not cancel their scores rather than believe that their 160 may have been a 150 if...

You have a big, yet very common, misconception about how the LSAT is scored. The LSAT is NOT graded on a curve. Scaled scores achieved by people that don't cancel ARE NOT curved. How the population of people that took the same test as you performed DOES NOT determine the scoring scale. There is a big difference between a test being graded on a conversion scale and a test being graded on a curve.

The LSAT is a standardized test. 150, 160, 165, 170, etc. has to mean the same thing every time (an ability/performance level) and be comparable across administrations/test forms, no matter which test it is from and no matter whether the group of people that took that test overall were a bunch of geniuses or a bunch of dim wits.

Other people canceling has no effect on your scaled score.

So how does the LSAC make a scale in advance? Does it come from non-disclosed Feb test? I understand it is not curved but I do not know why certain exams are 10 for 170, others 12 for 170. Is it just the matter of difficulty assumed by the LSAC?

Surprisingly, the experts over at LSAC have done this a time or two before. When you're dealing with a population of 40,000+ and you have used these questions in previous experimental sections, the distribution ends up being very predictable. However, if a question comes back with a very different distribution than they anticipated, it is thrown out. That is how questions are removed.

Why would LSAC need to "assume" difficulty when the organization routinely administers experimental sections under actual testing conditions?

Doubt it will have any impact