High scorers: do you perform well on every diagnostic test?

Prepare for the LSAT or discuss it with others in this forum.
Post Reply

Did your raw scores fluctuate, and how difficulty did you find the practice tests?

My raw score did not fluctuate much, and I found tests were about the same difficulty
5
50%
My raw score did not fluctuate much, but I found that tests noticeably varied in difficulty
1
10%
My raw score fluctuated a lot, but I found that tests were about the same difficulty
0
No votes
My raw score fluctuated a lot, and I found that tests noticeably varied in difficulty
4
40%
 
Total votes: 10

mandrewsf

New
Posts: 24
Joined: Sun Oct 19, 2014 6:01 pm

High scorers: do you perform well on every diagnostic test?

Post by mandrewsf » Thu Aug 06, 2020 6:39 pm

Having done 10 full-length practice tests so far, I'm seeing some fluctuation on my test results, particularly my raw score which can differ between 5-6 from test to test. Now, I know people will say that what matters is the reported score rather than how many raw questions you get right, but given the way LSAT is graded, the reported score you get depends on the ability of your fellow test-takers, while the raw score is a factor that the test taker could fully control. Hence, it appears that being able to get a good raw score on every practice tests should be more valuable than getting a good reported score on every practice test.

So, to those of you who scored 172+ on your LSAT/typically getting 172+ on your practice tests, when you practiced, did you typically get the same number of raw questions right on each one of your practice tests? Furthermore, did you find each of your practice tests to be of roughly the same difficulty, or did you find tests (not just individual sections, but overall) to noticeably fluctuate in difficulty?

replevin123

New
Posts: 24
Joined: Sun Jun 02, 2019 2:50 am

Re: High scorers: do you perform well on every diagnostic test?

Post by replevin123 » Thu Aug 06, 2020 7:16 pm

a little hard to remember back to those days, but I believe my raw scores fluctuated 2-3 from the mean. I was a 175+ scorer and was PT'ing consistently in the 173-180 range. On any given test, I was generally missing 0-1 on LG (missing one would've been bad for me); 1-4 on LR; and 1-2 on RC. Something like that, give or take

User avatar
Pneumonia

Gold
Posts: 1975
Joined: Sun Jul 29, 2012 3:05 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by Pneumonia » Thu Aug 06, 2020 8:08 pm

replevin123 wrote:
Thu Aug 06, 2020 7:16 pm
a little hard to remember back to those days, but I believe my raw scores fluctuated 2-3 from the mean. I was a 175+ scorer and was PT'ing consistently in the 173-180 range. On any given test, I was generally missing 0-1 on LG (missing one would've been bad for me); 1-4 on LR; and 1-2 on RC. Something like that, give or take
Same here, though perhaps fluctuating 3-4.

OP, you understand the difference between scaled and curved, right? Your score depends on past test takers, not your fellow (current) test takers. For that reason, I think you should focus more on the scaled score.

HopefulSplitter0000

Bronze
Posts: 116
Joined: Mon Apr 30, 2018 8:51 am

Re: High scorers: do you perform well on every diagnostic test?

Post by HopefulSplitter0000 » Thu Aug 06, 2020 9:39 pm

177 LSAT

I consistently missed 0 on logic games, 0-1 on logical reasoning, and 1-4 on reading comprehension. My highest practice test was 180 (scored maybe 3-4 times), but my scores were all above 174 by the time I took my last 10 practice tests before the exam.

Now, to address another point: I disagree that 5-6 points is a large amount/fluctuation. Once you pass 172 or so, the reading comprehension section alone, which is less formulaic than the others, can cause 3-5 point fluctuations.

Moreover, It is not only a bad idea to look at raw score, but also to look at total score. If your score on LG or LR is changing by more than 1-2 points, you are not ready. These sections are the same on every exam, and the fact that the test-makers choose new words should not impact difficulty. That said, for reading comprehension, if you are seeing 3-4 point fluctuations, as long as your score is on the high end, I would call that normal.

ksm6969

New
Posts: 38
Joined: Sun Jan 19, 2020 11:28 am

Re: High scorers: do you perform well on every diagnostic test?

Post by ksm6969 » Thu Aug 06, 2020 9:50 pm

HopefulSplitter0000 wrote:
Thu Aug 06, 2020 9:39 pm
177 LSAT

I consistently missed 0 on logic games, 0-1 on logical reasoning, and 1-4 on reading comprehension. My highest practice test was 180 (scored maybe 3-4 times), but my scores were all above 174 by the time I took my last 10 practice tests before the exam.

Now, to address another point: I disagree that 5-6 points is a large amount/fluctuation. Once you pass 172 or so, the reading comprehension section alone, which is less formulaic than the others, can cause 3-5 point fluctuations.

Moreover, It is not only a bad idea to look at raw score, but also to look at total score. If your score on LG or LR is changing by more than 1-2 points, you are not ready. These sections are the same on every exam, and the fact that the test-makers choose new words should not impact difficulty. That said, for reading comprehension, if you are seeing 3-4 point fluctuations, as long as your score is on the high end, I would call that normal.
Just wanted to +1 this whole thing. It’s exactly right.

Want to continue reading?

Register now to search topics and post comments!

Absolutely FREE!


mandrewsf

New
Posts: 24
Joined: Sun Oct 19, 2014 6:01 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by mandrewsf » Thu Aug 06, 2020 11:18 pm

Pneumonia wrote:
Thu Aug 06, 2020 8:08 pm
replevin123 wrote:
Thu Aug 06, 2020 7:16 pm
a little hard to remember back to those days, but I believe my raw scores fluctuated 2-3 from the mean. I was a 175+ scorer and was PT'ing consistently in the 173-180 range. On any given test, I was generally missing 0-1 on LG (missing one would've been bad for me); 1-4 on LR; and 1-2 on RC. Something like that, give or take
Same here, though perhaps fluctuating 3-4.

OP, you understand the difference between scaled and curved, right? Your score depends on past test takers, not your fellow (current) test takers. For that reason, I think you should focus more on the scaled score.
Yeah, which is to say the scaled score of the past test takers are not indicative of your actual performance on that practice test, since you're competing with a different group of test takers. Hence raw score being more indicative of your actual level of performance.

User avatar
Pneumonia

Gold
Posts: 1975
Joined: Sun Jul 29, 2012 3:05 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by Pneumonia » Thu Aug 06, 2020 11:24 pm

mandrewsf wrote:
Thu Aug 06, 2020 11:18 pm
Hence raw score being more indicative of your actual level of performance.
No, the opposite. A -13 test is objectively harder than a -9 test. If you get a 170 on both, your level of performance was the same even though your raw score was four questions different. At least that is how it is supposed to be. Scaling normalizes between tests, not groups of test takers.

mandrewsf

New
Posts: 24
Joined: Sun Oct 19, 2014 6:01 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by mandrewsf » Thu Aug 06, 2020 11:29 pm

HopefulSplitter0000 wrote:
Thu Aug 06, 2020 9:39 pm
177 LSAT

I consistently missed 0 on logic games, 0-1 on logical reasoning, and 1-4 on reading comprehension. My highest practice test was 180 (scored maybe 3-4 times), but my scores were all above 174 by the time I took my last 10 practice tests before the exam.

Now, to address another point: I disagree that 5-6 points is a large amount/fluctuation. Once you pass 172 or so, the reading comprehension section alone, which is less formulaic than the others, can cause 3-5 point fluctuations.

Moreover, It is not only a bad idea to look at raw score, but also to look at total score. If your score on LG or LR is changing by more than 1-2 points, you are not ready. These sections are the same on every exam, and the fact that the test-makers choose new words should not impact difficulty. That said, for reading comprehension, if you are seeing 3-4 point fluctuations, as long as your score is on the high end, I would call that normal.
Thanks for your response. So instead of looking at either the raw or the scaled score, a better measurement is the level of fluctuation for the LG and LR sections? That does make a lot of intuitive sense. I'm not at the point where I could say LR is formulaic yet, at least compared to LG, but perhaps that's the direction I should be working toward.

mandrewsf

New
Posts: 24
Joined: Sun Oct 19, 2014 6:01 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by mandrewsf » Thu Aug 06, 2020 11:34 pm

Pneumonia wrote:
Thu Aug 06, 2020 11:24 pm
mandrewsf wrote:
Thu Aug 06, 2020 11:18 pm
Hence raw score being more indicative of your actual level of performance.
No, the opposite. A -13 test is objectively harder than a -9 test. If you get a 170 on both, your level of performance was the same even though your raw score was four questions different. At least that is how it is supposed to be. Scaling normalizes between tests, not groups of test takers.
It's been my understanding that the scaled score depends on the percentage of test takers who scored certain raw scores, with LSAC distributing those raw scores on the 180-point scale. As a result, those who score 170 and above typically counts for 2% of the top test takers, those who score 172 and above counts for 1% of the top test takers, and so on. So are you saying that my understanding is incorrect, and the scaled score of an LSAT is determined in advance, without considering the results of the test takers?

Want to continue reading?

Register for access!

Did I mention it was FREE ?


User avatar
Pneumonia

Gold
Posts: 1975
Joined: Sun Jul 29, 2012 3:05 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by Pneumonia » Thu Aug 06, 2020 11:46 pm

That’s right, the scaled score for a test is set before it is administered. What you described is a curve. The LSAT is scaled before the test is given, not curved after it’s been graded. The scale aims to achieve the distribution you described, but in a given sitting, it is possible for every test taker to get 170+ or for no taker to get 170+. That is because the scale is set before anyone actually sits for the test.

FWIW I agree with the poster who said you should be tracking section by section rather than raw scores overall. Consistent -0/-1 on LG is a must. Every section tests the same skill, just with increasing levels of obfuscation. LG spoon feeds you the rules, LR mask them a bit, and RC masks them a lot.

mandrewsf

New
Posts: 24
Joined: Sun Oct 19, 2014 6:01 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by mandrewsf » Thu Aug 06, 2020 11:56 pm

Pneumonia wrote:
Thu Aug 06, 2020 11:46 pm
That’s right, the scaled score for a test is set before it is administered. What you described is a curve. The LSAT is scaled before the test is given, not curved after it’s been graded. The scale aims to achieve the distribution you described, but in a given sitting, it is possible for every test taker to get 170+ or for no taker to get 170+. That is because the scale is set before anyone actually sits for the test.

FWIW I agree with the poster who said you should be tracking section by section rather than raw scores overall. Consistent -0/-1 on LG is a must. Every section tests the same skill, just with increasing levels of obfuscation. LG spoon feeds you the rules, LR mask them a bit, and RC masks them a lot.
This is actually great to know, thanks. It turns out the LSAC test writers are even better at their jobs than I'd thought. Appreciate the helpful study advice too.

User avatar
awprufrock

New
Posts: 2
Joined: Sun Nov 03, 2019 6:25 pm

Re: High scorers: do you perform well on every diagnostic test?

Post by awprufrock » Fri Aug 07, 2020 10:28 pm

Pneumonia wrote:
Thu Aug 06, 2020 11:46 pm
That’s right, the scaled score for a test is set before it is administered. What you described is a curve. The LSAT is scaled before the test is given, not curved after it’s been graded. The scale aims to achieve the distribution you described, but in a given sitting, it is possible for every test taker to get 170+ or for no taker to get 170+. That is because the scale is set before anyone actually sits for the test.

FWIW I agree with the poster who said you should be tracking section by section rather than raw scores overall. Consistent -0/-1 on LG is a must. Every section tests the same skill, just with increasing levels of obfuscation. LG spoon feeds you the rules, LR mask them a bit, and RC masks them a lot.
This is really helpful, appreciate it.

Register now!

Resources to assist law school applicants, students & graduates.

It's still FREE!


Post Reply

Return to “LSAT Prep and Discussion Forum”