What actually happens to these undisclosed February tests? Forum

[justbubbles]

Does LSAC recycle these questions (with slightly different subjects) on subsequent tests?

Does LSAC give these tests to non-traditional test-takers (eg - overseas LSAT seating)?

Does LSAC just use the February tests to run a huge guinea pig experiment as in one huge test with 5 experimental sections?

Anyone ever see a February test question (or the like) appear on a disclosed (future) test?

Just curious.

[shempskidd]

I think questions one and two supply the best and most truthful answers.

[LSAT Blog]

LSAC says they need to have nondisclosed tests "in reserve for emergencies and special uses."

I take this to mean that they need to have nondisclosed tests for use in case of reschedulings due to inclement weather, for use in overseas administrations (those outside the Americas), and for Sabbath observers' administrations.

Details: http://lsatblog.blogspot.com/2011/12/wh ... losed.html

[justbubbles]

Thanks, Steve.

I don't have issues with this test being undisclosed for the reasons that they have specified, but I wish LSAC would amend their policy on two things about the February test which are:

1. Disclosure of score-by-section breakdown - If LSAC would just disclose who scores what in what section, I don't see how their reasons for "non-disclosure" would be compromised.

2. Concession in test-taking fees - It seems to me that it is inherently unfair for test-takers to be paying the same amount for the February test where they don't get to see their answers and they don't get to keep the test-questions and answers. (I took the test on a fee waiver. I am not complaining, but I am just saying to be fair to everyone else).

[LSAT Blog]

I could reasonably see #1 happening someday, but I wouldn't be so optimistic about #2.

[bernaldiaz]

Thanks, Steve.

I don't have issues with this test being undisclosed for the reasons that they have specified, but I wish LSAC would amend their policy on two things about the February test which are:

1. Disclosure of score-by-section breakdown - If LSAC would just disclose who scores what in what section, I don't see how their reasons for "non-disclosure" would be compromised.

2. Concession in test-taking fees - It seems to me that it is inherently unfair for test-takers to be paying the same amount for the February test where they don't get to see their answers and they don't get to keep the test-questions and answers. (I took the test on a fee waiver. I am not complaining, but I am just saying to be fair to everyone else).

Yeah, if they'd just let us see our Scantron that would take care of that. I'm a bit worried because I had to erase a few questions that the machine scoring will mark a couple wrong, so I'll probably dole out the 42 bucks now to contest the score no matter what.

[Clearly]

Eh, I understand the motivations, but if not fr the restriction on posting answers, I can recall at least 60% of the test and reproduce it. I've already written all four logic games, And the rules down for each of them, I know all the reading comps, and could find the full passages, and about half of the LR questions.
I wouldn't share this information, but if I did some sabbath observer would be very happy...

The point is, this isn't a safe method for withholding questions for other testers, while I wouldn't share what I know about the test, I'm certain not everyone else would uphold the commitment not to share test materials.

[bernaldiaz]

Eh, I understand the motivations, but if not fr the restriction on posting answers, I can recall at least 60% of the test and reproduce it. I've already written all four logic games, And the rules down for each of them, I know all the reading comps, and could find the full passages, and about half of the LR questions.
I wouldn't share this information, but if I did some sabbath observer would be very happy...

The point is, this isn't a safe method for withholding questions for other testers, while I wouldn't share what I know about the test, I'm certain not everyone else would uphold the commitment not to share test materials.

They get a different test. And we wouldn't get the results until weeks after they took it.

[justbubbles]

Yeah, if they'd just let us see our Scantron that would take care of that. I'm a bit worried because I had to erase a few questions that the machine scoring will mark a couple wrong, so I'll probably dole out the 42 bucks now to contest the score no matter what.

Yep, same here. I amended quite a few of my answers post-bubbling. We all know Scantron readings are not fully immune from errors and omissions.

[maxmartin]

Does LSAC just use the February tests to run a huge guinea pig experiment as in one huge test with 5 experimental sections?

Yes I think this is also quite possible. there are already multiple people confirmed their LR and LR sections are different from majority people. So I am not surprised undisclosed itself is a huge experiment for measure the test quality and future LSAT test. Think about those people who received unusual tests as internal standards, who knows how they will be scored. In a disclosed test, you won't have have any internal standard.

Thanks, Steve.

I don't have issues with this test being undisclosed for the reasons that they have specified, but I wish LSAC would amend their policy on two things about the February test which are:

1. Disclosure of score-by-section breakdown - If LSAC would just disclose who scores what in what section, I don't see how their reasons for "non-disclosure" would be compromised.

2. Concession in test-taking fees - It seems to me that it is inherently unfair for test-takers to be paying the same amount for the February test where they don't get to see their answers and they don't get to keep the test-questions and answers. (I took the test on a fee waiver. I am not complaining, but I am just saying to be fair to everyone else).

[suspicious android]

Yes I think this is also quite possible. there are already multiple people confirmed their LR and LR sections are different from majority people. So I am not surprised undisclosed itself is a huge experiment for measure the test quality and future LSAT test. Think about those people who received unusual tests as internal standards, who knows how they will be scored. In a disclosed test, you won't have have any internal standard.

No. The test is standardized. That's the whole point of the test.

[Jeffort]

Yeah, if they'd just let us see our Scantron that would take care of that. I'm a bit worried because I had to erase a few questions that the machine scoring will mark a couple wrong, so I'll probably dole out the 42 bucks now to contest the score no matter what.

The machines they use to scan the answer sheets are very good and don't make errors due a little bit of residue as long you erased the entire oval. They differentiate without any trouble a fully filled in bubble from another in the same row with some graphite smug residual due to having erased the oval.

If you submit a hand-score request LSAC wants you to specify which section and question number(s) you think might have been read incorrectly by the machine. As long as you filled in your selected answer choice oval properly and adequately erased the other choice, it is highly unlikely that your answer sheet will be scored incorrectly. I've never heard of a case of that happening.

If you didn't fill in your selected answer choice oval fully or at least close to fully and/or didn't really erase another bubble (meaning you left a darkened area that was not rubbed out much or at all), then you are out of luck and hand scoring will not help you.

Many hand scoring issues/problems that pop-up and sometimes get remedied are for miss-gridding a bunch of consecutive questions in the same section because of accidentally skipping a row so that a chunk of answers are offset by one row number. Another one is due to people that just didn't fill in an oval enough, but they don't tend to be very forgiving for that type of mistake.

The gray residue smudge from erasing a bubble is not a problem. When you erase #2/HB pencil graphite, it always leaves a light visible gray smudge on the paper unless you bleach it, use white-out or some other cover it up stuff.

[bernaldiaz]

Yeah, if they'd just let us see our Scantron that would take care of that. I'm a bit worried because I had to erase a few questions that the machine scoring will mark a couple wrong, so I'll probably dole out the 42 bucks now to contest the score no matter what.

The machines they use to scan the answer sheets are very good and don't make errors due a little bit of residue as long you erased the entire oval. They differentiate without any trouble a fully filled in bubble from another in the same row with some graphite smug residual due to having erased the oval.

If you submit a hand-score request LSAC wants you to specify which section and question number(s) you think might have been read incorrectly by the machine. As long as you filled in your selected answer choice oval properly and adequately erased the other choice, it is highly unlikely that your answer sheet will be scored incorrectly. I've never heard of a case of that happening.

If you didn't fill in your selected answer choice oval fully or at least close to fully and/or didn't really erase another bubble (meaning you left a darkened area that was not rubbed out much or at all), then you are out of luck and hand scoring will not help you.

Many hand scoring issues/problems that pop-up and sometimes get remedied are for miss-gridding a bunch of consecutive questions in the same section because of accidentally skipping a row so that a chunk of answers are offset by one row number. Another one is due to people that just didn't fill in an oval enough, but they don't tend to be very forgiving for that type of mistake.

The gray residue smudge from erasing a bubble is not a problem. When you erase #2/HB pencil graphite, it always leaves a light visible gray smudge on the paper unless you bleach it, use white-out or some other cover it up stuff.

Thanks Jeffort, very helpful.