The "Ask @JoeSeperac" Thread

Discussions related to the bar exam are found in this forum
JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Fri Mar 15, 2019 10:58 am

John--- wrote:Joe (or anyone else who might know the answer): The maximum scaled score I can enter in your calculator for MEE is a number around 67? And 85 for MPT? https://seperac.com/zcalc-ube-f18.php

Extreme values can make the calculator go haywire. High scoring examinees generally pass so I never see their essay scores which would help me improve the calculator. On the UBE, the most you can receive for a single MEE answer is 20 UBE points (5% of 400), but with my F18 NY calculator, this occurs with a scaled essay score of 78. Meanwhile, I am aware of examinees who have scored higher than a 78 on the essays, so something else is going on that I don’t fully understand. For the NY UBE, a simple rule of thumb is a score of 20 if you wrote nothing and a score of 80 for a top 99th percentile answer (although this vary by +/- 10).

Post-exam worrying about the essays is not productive. For example, the day before the July 2016 UBE exam, a subscriber (foreign examinee) emailed me the following: “You’re probably being swamped by panicking students today (including me), so I understand if you don’t have time to respond. Basically, I’m considering withdrawing from the exam and re-sitting in February because I feel very, very underprepared for the essays portion of the test.” I told this subscriber that it would be a mistake to withdraw (everyone feels unprepared for the exam). The examinee sat for the exam and ended up passing with a 177.9 scaled MEE/MPT score (MBE score of 148.1 and total score of 326). A written score of 177.9 places this examinee’s MEE/MPT scores in the 99.8% percentile among examinees nationwide (meaning just 0.2% of last year's examinees nationwide scored better than this examinee on the MEE/MPT). I later found out that this examinee wrote the 2nd released above average answer to Essay #4 (Secured Transactions), meaning this examinee wrote the highest scoring essay #4 out of 10,296 examinees. Being the highest scoring examinee on the Secured Transactions essay, this examinee received an estimated scaled score of 85 on that essay. This means the examinee needed to average 62 on the remaining essays/MPTs in order to score a 177.9 on the MEE/MPT. In contrast, the failing examines who sent me their J16 scores averaged 44 on the MEE questions and 43 on the MPT questions. So basically, a 180-185 is generally the most you can score on the MEEs/MPTs and MBE, meaning a maximum of 370/400 total score. To date, the highest score I’ve ever received from a UBE examinee is 355 (171 MBE, 184 written).

Bottom line, you never know what to expect on the written portion of the exam. Even if you feel “very, very underprepared for the essays portion of the test” as this examinee did, you can still score well on the essays/MPT. Sometimes it is luck (topics you had prepared for appeared) and sometimes it is unexplainable (essay grading is a very nebulous thing).

John---

New
Posts: 20
Joined: Fri Mar 01, 2019 1:21 pm

Re: The "Ask @JoeSeperac" Thread

Postby John--- » Fri Mar 15, 2019 7:38 pm

JoeSeperac wrote:
John--- wrote:Joe (or anyone else who might know the answer): The maximum scaled score I can enter in your calculator for MEE is a number around 67? And 85 for MPT? https://seperac.com/zcalc-ube-f18.php

Extreme values can make the calculator go haywire. High scoring examinees generally pass so I never see their essay scores which would help me improve the calculator. On the UBE, the most you can receive for a single MEE answer is 20 UBE points (5% of 400), but with my F18 NY calculator, this occurs with a scaled essay score of 78. Meanwhile, I am aware of examinees who have scored higher than a 78 on the essays, so something else is going on that I don’t fully understand. For the NY UBE, a simple rule of thumb is a score of 20 if you wrote nothing and a score of 80 for a top 99th percentile answer (although this vary by +/- 10).

Post-exam worrying about the essays is not productive. For example, the day before the July 2016 UBE exam, a subscriber (foreign examinee) emailed me the following: “You’re probably being swamped by panicking students today (including me), so I understand if you don’t have time to respond. Basically, I’m considering withdrawing from the exam and re-sitting in February because I feel very, very underprepared for the essays portion of the test.” I told this subscriber that it would be a mistake to withdraw (everyone feels unprepared for the exam). The examinee sat for the exam and ended up passing with a 177.9 scaled MEE/MPT score (MBE score of 148.1 and total score of 326). A written score of 177.9 places this examinee’s MEE/MPT scores in the 99.8% percentile among examinees nationwide (meaning just 0.2% of last year's examinees nationwide scored better than this examinee on the MEE/MPT). I later found out that this examinee wrote the 2nd released above average answer to Essay #4 (Secured Transactions), meaning this examinee wrote the highest scoring essay #4 out of 10,296 examinees. Being the highest scoring examinee on the Secured Transactions essay, this examinee received an estimated scaled score of 85 on that essay. This means the examinee needed to average 62 on the remaining essays/MPTs in order to score a 177.9 on the MEE/MPT. In contrast, the failing examines who sent me their J16 scores averaged 44 on the MEE questions and 43 on the MPT questions. So basically, a 180-185 is generally the most you can score on the MEEs/MPTs and MBE, meaning a maximum of 370/400 total score. To date, the highest score I’ve ever received from a UBE examinee is 355 (171 MBE, 184 written).

Bottom line, you never know what to expect on the written portion of the exam. Even if you feel “very, very underprepared for the essays portion of the test” as this examinee did, you can still score well on the essays/MPT. Sometimes it is luck (topics you had prepared for appeared) and sometimes it is unexplainable (essay grading is a very nebulous thing).



Thank you so much, Joe.

masonjarrr

New
Posts: 13
Joined: Sat Mar 09, 2019 1:22 am

Re: The "Ask @JoeSeperac" Thread

Postby masonjarrr » Fri Mar 15, 2019 10:52 pm

JoeSeperac wrote:
John--- wrote:Joe (or anyone else who might know the answer): The maximum scaled score I can enter in your calculator for MEE is a number around 67? And 85 for MPT? https://seperac.com/zcalc-ube-f18.php

Extreme values can make the calculator go haywire. High scoring examinees generally pass so I never see their essay scores which would help me improve the calculator. On the UBE, the most you can receive for a single MEE answer is 20 UBE points (5% of 400), but with my F18 NY calculator, this occurs with a scaled essay score of 78. Meanwhile, I am aware of examinees who have scored higher than a 78 on the essays, so something else is going on that I don’t fully understand. For the NY UBE, a simple rule of thumb is a score of 20 if you wrote nothing and a score of 80 for a top 99th percentile answer (although this vary by +/- 10).

Post-exam worrying about the essays is not productive. For example, the day before the July 2016 UBE exam, a subscriber (foreign examinee) emailed me the following: “You’re probably being swamped by panicking students today (including me), so I understand if you don’t have time to respond. Basically, I’m considering withdrawing from the exam and re-sitting in February because I feel very, very underprepared for the essays portion of the test.” I told this subscriber that it would be a mistake to withdraw (everyone feels unprepared for the exam). The examinee sat for the exam and ended up passing with a 177.9 scaled MEE/MPT score (MBE score of 148.1 and total score of 326). A written score of 177.9 places this examinee’s MEE/MPT scores in the 99.8% percentile among examinees nationwide (meaning just 0.2% of last year's examinees nationwide scored better than this examinee on the MEE/MPT). I later found out that this examinee wrote the 2nd released above average answer to Essay #4 (Secured Transactions), meaning this examinee wrote the highest scoring essay #4 out of 10,296 examinees. Being the highest scoring examinee on the Secured Transactions essay, this examinee received an estimated scaled score of 85 on that essay. This means the examinee needed to average 62 on the remaining essays/MPTs in order to score a 177.9 on the MEE/MPT. In contrast, the failing examines who sent me their J16 scores averaged 44 on the MEE questions and 43 on the MPT questions. So basically, a 180-185 is generally the most you can score on the MEEs/MPTs and MBE, meaning a maximum of 370/400 total score. To date, the highest score I’ve ever received from a UBE examinee is 355 (171 MBE, 184 written).

Bottom line, you never know what to expect on the written portion of the exam. Even if you feel “very, very underprepared for the essays portion of the test” as this examinee did, you can still score well on the essays/MPT. Sometimes it is luck (topics you had prepared for appeared) and sometimes it is unexplainable (essay grading is a very nebulous thing).


If this helps you for the future, I tried to see whether a extreme score made a magnified difference than a few similiarly situated 65s [you emailed me your assessment of my estimated mbe scores a week or two ago so very thankful for that]

Here's my scores from J18 [operant grades]
80
62.5
60
55
60
60 [PT]

If I were to bring that 55 in E4 to a 65, that 60 to a 65 in E3, and reduce that 60 to a 65, my total scaled score becomes 1430 from a 1428. So, it's better to do similarly well in all essays than do amazing in one and except a big jump.

paperplanes22

New
Posts: 8
Joined: Sat Feb 23, 2019 7:21 pm

Re: The "Ask @JoeSeperac" Thread

Postby paperplanes22 » Sat Mar 16, 2019 5:38 pm

Sorry if this has already been asked, but I looked and couldn’t find it if it has.

When I type the same raw scores into different states, let’s say 3s for all essays and MPTs and 138 for the MBE, I obviously get different scales scores, depending on the state. For example, in CO, this would be scaled to a 271 and in DC, to a 260. Does this mean that if I’m in a UBE state where the passing score is 260, I statistically have a better chance at hitting the 260 taking it in a state where the scaled score is higher?

Smiddywesson

Bronze
Posts: 164
Joined: Tue Feb 12, 2019 1:11 pm

Re: The "Ask @JoeSeperac" Thread

Postby Smiddywesson » Thu Mar 21, 2019 11:01 am

Hi Joe,

I encountered a question the other day that has plagued me, so I thought I'd ask the master. The post asked if the scale on the MBE could actually reduce your raw score based on group performance on the equators. I said no, but now I am in doubt. My initial response was based on the idea that the scale could be constructed so that a "reduction" for your group could be already built into the scale. For example, if your group performs equally as well as the previous February and July groups, your scale might begin at an 8, not at a zero. Therefore, a group that "loses" potential scaling points due to poor equator performance never sees points taken off their raw score, they just see less than 8 being added. The range of "negative" scale could be 7-0, but never reaches -1 and nobody is the wiser about what went into the secret sauce.

Am I off base, or is that how it works? I've never heard about people losing points off their raw score, but they generally don't release raw scores, do they? This tells me they don't want the peasants looking behind the curtain. Any enlightenment you could provide would be greatly appreciated.

JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Sun Mar 24, 2019 9:45 am

masonjarrr wrote:Here's my scores from J18 [operant grades]
80
62.5
60
55
60
60 [PT]

If I were to bring that 55 in E4 to a 65, that 60 to a 65 in E3, and reduce that 60 to a 65, my total scaled score becomes 1430 from a 1428. So, it's better to do similarly well in all essays than do amazing in one and except a big jump.


Can you double-check that? If I take your essay scores (using default 1440 for MBE) and then reduce E1 by 20 points and increase E2-E5 by 5 points each, I end up with the same score, so I'm really curious to see how you may have ended up with different scores

Image

JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Sun Mar 24, 2019 10:27 am

Smiddywesson wrote:Hi Joe,

I encountered a question the other day that has plagued me, so I thought I'd ask the master. The post asked if the scale on the MBE could actually reduce your raw score based on group performance on the equators. I said no, but now I am in doubt. My initial response was based on the idea that the scale could be constructed so that a "reduction" for your group could be already built into the scale. For example, if your group performs equally as well as the previous February and July groups, your scale might begin at an 8, not at a zero. Therefore, a group that "loses" potential scaling points due to poor equator performance never sees points taken off their raw score, they just see less than 8 being added. The range of "negative" scale could be 7-0, but never reaches -1 and nobody is the wiser about what went into the secret sauce.

Am I off base, or is that how it works? I've never heard about people losing points off their raw score, but they generally don't release raw scores, do they? This tells me they don't want the peasants looking behind the curtain. Any enlightenment you could provide would be greatly appreciated.


I discuss how raw scores are converted to scale scores here (along with a calculator):
http://www.seperac.com/zcalc-raw-scaled.php

In order to know the scale, you need to know the correct mean and standard deviation (SD) for the MBE in your jurisdiction for that exam, along with the MEEs/MPTs mean and SD. I don’t know of any jurisdiction that divulges this information anymore (PA bar exam once reported this information in 2002 and that’s it).

If you play around with the calculator and put in a high Essay mean raw score but low Essay raw SD, it will result in a negative scale. Whether this has ever happened in a bar administration, I haven't a clue.

What state did you take the exam in? I can let you know their average pass rates over the past 20+ years. That will give you a little more insight into your odds of passing.

JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Sun Mar 24, 2019 10:28 am

paperplanes22 wrote:Sorry if this has already been asked, but I looked and couldn’t find it if it has.

When I type the same raw scores into different states, let’s say 3s for all essays and MPTs and 138 for the MBE, I obviously get different scales scores, depending on the state. For example, in CO, this would be scaled to a 271 and in DC, to a 260. Does this mean that if I’m in a UBE state where the passing score is 260, I statistically have a better chance at hitting the 260 taking it in a state where the scaled score is higher?


Yes, that seems to be the case and the evidence supports it. Both DC and CO have the same essay grading scale of 1-6 so the difference can't be attributed to the grading scale.

In District of Columbia over the past 23 years, the February ABA First-Timers Pass Rate is 56.8% while the July ABA First-Timers Pass Rate is 76.3%. The February ABA Repeaters Pass Rate is 30.4% while the July ABA Repeaters Pass Rate is 29.4%.

In Colorado over the past 23 years, the February ABA First-Timers Pass Rate is 75.8% while the July ABA First-Timers Pass Rate is 83.1%. The February ABA Repeaters Pass Rate is 44.7% while the July ABA Repeaters Pass Rate is 31.5%.

An issue you face is that state pass rates may change to prevent forum shopping. For example, NJ had an overall July pass rate of 75% based on the last 20 years while CT had an overall July pass rate of 77% based on the last 20 years. The J18 NJ pass rate was 59% (drop of 16%) while the J18 CT pass rate was 55% (drop of 22%). Personally I think this was done to stop failing New Yorkers from UBE forum shopping (all 3 states are 266).

As an aside, if you failed the exam and send me your scores, I will give you a free comprehensive analysis of your scores and it helps me me bar score calculators for other states.

https://www.seperac.com/scoreform.php

masonjarrr

New
Posts: 13
Joined: Sat Mar 09, 2019 1:22 am

Re: The "Ask @JoeSeperac" Thread

Postby masonjarrr » Sun Mar 24, 2019 5:24 pm

JoeSeperac wrote:
masonjarrr wrote:Here's my scores from J18 [operant grades]
80
62.5
60
55
60
60 [PT]

If I were to bring that 55 in E4 to a 65, that 60 to a 65 in E3, and reduce that 60 to a 65, my total scaled score becomes 1430 from a 1428. So, it's better to do similarly well in all essays than do amazing in one and except a big jump.


Can you double-check that? If I take your essay scores (using default 1440 for MBE) and then reduce E1 by 20 points and increase E2-E5 by 5 points each, I end up with the same score, so I'm really curious to see how you may have ended up with different scores


I received a 1377 as my scaled MBE score. I tried it again, but I think I may have inputted the wrong numbers. I am getting the same 1428.03 by changing up the essay scores now. I guess the raw written is the final determinative score regardless of the variations between the individual essay scores.

I had (may be) read up at one of the official sites that a significantly higher score in one may sway the score to your favor, but that's doesn't seem to be the case. This inquiry was a nothing burger lol

TheJuryMustDie

Bronze
Posts: 164
Joined: Mon Mar 20, 2017 12:27 pm

Re: The "Ask @JoeSeperac" Thread

Postby TheJuryMustDie » Mon Mar 25, 2019 9:15 am

Ha! I reckon that's not possible. Primarily because it will undermine the MBE scale being expressed in 200. Unless of course, the 200 scale is just a ruse and the NCBE doesn't really scale raw scores up to 200 as they claim.

If we believe, as we are expected to, that the MBE scale is 200, then each raw score/mark/point is worth circa 1.15 scaled points. While I believe for statistical purposes some questions (especially the equators) are more important to the NCBE than other questions, I think it will be greatly arbitral and perhaps misleading to add more point weight to some questions than others.

And more importantly, if that was the case, the big prep companies would have been on it too.

Smiddywesson wrote:Hi Joe,

I encountered a question the other day that has plagued me, so I thought I'd ask the master. The post asked if the scale on the MBE could actually reduce your raw score based on group performance on the equators. I said no, but now I am in doubt. My initial response was based on the idea that the scale could be constructed so that a "reduction" for your group could be already built into the scale. For example, if your group performs equally as well as the previous February and July groups, your scale might begin at an 8, not at a zero. Therefore, a group that "loses" potential scaling points due to poor equator performance never sees points taken off their raw score, they just see less than 8 being added. The range of "negative" scale could be 7-0, but never reaches -1 and nobody is the wiser about what went into the secret sauce.

Am I off base, or is that how it works? I've never heard about people losing points off their raw score, but they generally don't release raw scores, do they? This tells me they don't want the peasants looking behind the curtain. Any enlightenment you could provide would be greatly appreciated.

lawgirl3521

New
Posts: 33
Joined: Thu Feb 28, 2019 12:11 pm

Re: The "Ask @JoeSeperac" Thread

Postby lawgirl3521 » Wed Mar 27, 2019 9:50 am

My apologies if this has already been asked - how does essay grading / scaling work if everyone does really well on essays? From what I have read and heard, pretty much everyone thought the essays were really straightforward, if not easy. Does that mean it becomes harder to pass on the essays, or can everyone actually pass? I am not a stats person *at all*, so please excuse the elementary nature of the question.

Also, my state (MO) does not give us a breakdown of any of our essays. We just get two scores: a scaled essay score where the MPT and MEE is combined, and an MBE score. So if your essay score is low, there is no way to determine your weak areas.

Anon-non-anon

New
Posts: 11
Joined: Fri Mar 15, 2019 12:40 pm

Re: The "Ask @JoeSeperac" Thread

Postby Anon-non-anon » Wed Mar 27, 2019 1:16 pm

Hi Joe,

Way late but just came across this thread and now I'm very curious. I took the July 2016 bar in NY and got a total NY bar score of 341 and scaled MBE of 175.2. Where does that put me?

Thanks!

Smiddywesson

Bronze
Posts: 164
Joined: Tue Feb 12, 2019 1:11 pm

Re: The "Ask @JoeSeperac" Thread

Postby Smiddywesson » Wed Mar 27, 2019 2:29 pm

lawgirl3521 wrote:My apologies if this has already been asked - how does essay grading / scaling work if everyone does really well on essays? From what I have read and heard, pretty much everyone thought the essays were really straightforward, if not easy. Does that mean it becomes harder to pass on the essays, or can everyone actually pass? I am not a stats person *at all*, so please excuse the elementary nature of the question.

Also, my state (MO) does not give us a breakdown of any of our essays. We just get two scores: a scaled essay score where the MPT and MEE is combined, and an MBE score. So if your essay score is low, there is no way to determine your weak areas.


Correct me if I'm wrong anybody, but the way I think it works is a bit like musical chairs. The essays are graded, and the MBE is graded. Then they scale the essays to the MBE by using those MBE scores like chairs in the game musical chairs. The highest graded essays are awarded the points earned by the highest MBE scores. For example, if the highest MBE score is a 178, then the highest grades essays will be awarded a 178.

This is why it's so important to focus on the MBE where your life is in your hands. On the other hand, not only are the initial grading of the essays subjective, but there are only so many good chairs (MBE scores) and if they are all taken by the time your essays come to the front of the line, you are screwed.

I10attorney

New
Posts: 14
Joined: Wed Oct 24, 2018 8:19 am

Re: The "Ask @JoeSeperac" Thread

Postby I10attorney » Fri Apr 05, 2019 6:14 am

Good morning, Joe. First off, thanks for all you do to keep countless folks semi-sane in the months following the exam. The wait is far worse than the preparation. A few states are releasing scores today. I (and surely hoards of others) am waiting with baited breath for your opinion regarding this administration's mean. Hopefully, your magic machine tells us the scale was more favorable than Feb 2018 :) .

lawgirl3521

New
Posts: 33
Joined: Thu Feb 28, 2019 12:11 pm

Re: The "Ask @JoeSeperac" Thread

Postby lawgirl3521 » Fri Apr 05, 2019 10:52 am

Smiddywesson wrote:
lawgirl3521 wrote:My apologies if this has already been asked - how does essay grading / scaling work if everyone does really well on essays? From what I have read and heard, pretty much everyone thought the essays were really straightforward, if not easy. Does that mean it becomes harder to pass on the essays, or can everyone actually pass? I am not a stats person *at all*, so please excuse the elementary nature of the question.

Also, my state (MO) does not give us a breakdown of any of our essays. We just get two scores: a scaled essay score where the MPT and MEE is combined, and an MBE score. So if your essay score is low, there is no way to determine your weak areas.


Correct me if I'm wrong anybody, but the way I think it works is a bit like musical chairs. The essays are graded, and the MBE is graded. Then they scale the essays to the MBE by using those MBE scores like chairs in the game musical chairs. The highest graded essays are awarded the points earned by the highest MBE scores. For example, if the highest MBE score is a 178, then the highest grades essays will be awarded a 178.

This is why it's so important to focus on the MBE where your life is in your hands. On the other hand, not only are the initial grading of the essays subjective, but there are only so many good chairs (MBE scores) and if they are all taken by the time your essays come to the front of the line, you are screwed.


I think I understand what you are saying? But, for example in MO, they release the highest scores for both essays and MBE and the numbers are not the same - would you mind elaborating? I don't know why these essays are bothering me so much. Here is a link to what I am talking about: https://www.mble.org/exam-web-stats-july-2018

JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Fri Apr 05, 2019 3:04 pm

lawgirl3521 wrote:My apologies if this has already been asked - how does essay grading / scaling work if everyone does really well on essays? From what I have read and heard, pretty much everyone thought the essays were really straightforward, if not easy. Does that mean it becomes harder to pass on the essays, or can everyone actually pass? I am not a stats person *at all*, so please excuse the elementary nature of the question.

Also, my state (MO) does not give us a breakdown of any of our essays. We just get two scores: a scaled essay score where the MPT and MEE is combined, and an MBE score. So if your essay score is low, there is no way to determine your weak areas.

I wish I could answer your question, but I can't. Essay scaling is a very nebulous thing. For example, I honestly think there were two distinct and separate essay scales for the J18 Massachusetts UBE (one for retakers and one for first-time examinees - I would love to get more scores from failing J18 MA examinees to be able to prove this). If you give me enough scores, I can figure out the scale, but as to what criteria is used to determine raw scores, your guess is as good as mine. For example, on the J10 NY pre-UBE exam, Question #3 was a Corporations essay which was rather difficult. Based on the essays I received from 57 examinees, the average answer length for this essay was 542 words with the shortest passing answer to Essay #3 being 347 words long. This is extremely short as compared to normal essay answer at the time. However, for each essay, the bar examiners determine the order of the examinees based on the quality of their answers and then rank them accordingly, which in this case led to very short answers becoming passing essays.

JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Fri Apr 05, 2019 3:12 pm

Anon-non-anon wrote:Hi Joe,

Way late but just came across this thread and now I'm very curious. I took the July 2016 bar in NY and got a total NY bar score of 341 and scaled MBE of 175.2. Where does that put me?

Thanks!

Sorry for the delayed follow up. Based on your scaled MBE score of 175.2, your estimated raw MBE score was about 164/175 correct (based on my estimate of the J16 MBE scale). This means you answered about 93.7% of the graded MBE questions correctly. This places you in the 99.1% percentile for the MBE. This means that 0.9% of Jul examinees nationwide did better than you on the MBE based on your scaled MBE score of 175.2 (based on July national data for the past 7 years).

Based on a total score of 341, this means your written score was 165.8, which would have placed you in the 93.8% percentile among examinees nationwide (meaning that 6.2% of examinees nationwide would have scored better than you on the MEE/MPT).

If you don't mind (and can still remember), how many questions did you answer in practice, from what sources (e.g. Barbri, Kaplan, Adaptibar, NCBE) and what was your overall % correct?

Joe

JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Fri Apr 05, 2019 3:14 pm

I10attorney wrote:Good morning, Joe. First off, thanks for all you do to keep countless folks semi-sane in the months following the exam. The wait is far worse than the preparation. A few states are releasing scores today. I (and surely hoards of others) am waiting with baited breath for your opinion regarding this administration's mean. Hopefully, your magic machine tells us the scale was more favorable than Feb 2018 :) .

In looking at the Feb 2017 and Feb 2018 Illinois MBE percentiles, I interpolated a national scaled MBE mean of 133.4 for Feb 2019. Although I think I am pretty close, please keep in mind that THIS IS JUST A GUESS. If I am correct, then the MBE mean went up from the Feb 2018 national MBE mean of 132.8 but it would still be the 2nd lowest Feb MBE mean since the MBE was introduced in 1972 (with F18 being the lowest).

Smiddywesson

Bronze
Posts: 164
Joined: Tue Feb 12, 2019 1:11 pm

Re: The "Ask @JoeSeperac" Thread

Postby Smiddywesson » Fri Apr 05, 2019 4:03 pm

lawgirl3521 wrote:
Smiddywesson wrote:
lawgirl3521 wrote:My apologies if this has already been asked - how does essay grading / scaling work if everyone does really well on essays? From what I have read and heard, pretty much everyone thought the essays were really straightforward, if not easy. Does that mean it becomes harder to pass on the essays, or can everyone actually pass? I am not a stats person *at all*, so please excuse the elementary nature of the question.

Also, my state (MO) does not give us a breakdown of any of our essays. We just get two scores: a scaled essay score where the MPT and MEE is combined, and an MBE score. So if your essay score is low, there is no way to determine your weak areas.


Correct me if I'm wrong anybody, but the way I think it works is a bit like musical chairs. The essays are graded, and the MBE is graded. Then they scale the essays to the MBE by using those MBE scores like chairs in the game musical chairs. The highest graded essays are awarded the points earned by the highest MBE scores. For example, if the highest MBE score is a 178, then the highest grades essays will be awarded a 178.

This is why it's so important to focus on the MBE where your life is in your hands. On the other hand, not only are the initial grading of the essays subjective, but there are only so many good chairs (MBE scores) and if they are all taken by the time your essays come to the front of the line, you are screwed.


I think I understand what you are saying? But, for example in MO, they release the highest scores for both essays and MBE and the numbers are not the same - would you mind elaborating? I don't know why these essays are bothering me so much. Here is a link to what I am talking about: https://www.mble.org/exam-web-stats-july-2018


I haven't a clue. If what I read was true, the highest essay should match the highest MBE score. We're missing part of the puzzle. As for the lowest scores, if you don't put down anything on your essays, you don't get the lowest MBE score, you get a zero, so that makes sense to me. Maybe Joe can explain this, there's some sort of secret sauce going on here, and I strongly suspect there's a lost of subjectivity from state to state. Maybe if you badmouth the Crimson Tide on your Mississippi bar, you get bonus points?

JoeSeperac

Moderator
Posts: 373
Joined: Thu Feb 16, 2017 3:30 pm

Re: The "Ask @JoeSeperac" Thread

Postby JoeSeperac » Sat Apr 06, 2019 10:15 am

Smiddywesson wrote:I haven't a clue. If what I read was true, the highest essay should match the highest MBE score. We're missing part of the puzzle. As for the lowest scores, if you don't put down anything on your essays, you don't get the lowest MBE score, you get a zero, so that makes sense to me. Maybe Joe can explain this, there's some sort of secret sauce going on here, and I strongly suspect there's a lost of subjectivity from state to state. Maybe if you badmouth the Crimson Tide on your Mississippi bar, you get bonus points?

In New York, even if you write nothing for an essay/MPT, you will still receive a scaled score that can range from 5-25. For states like Mississippi, I haven't a clue. You are correct that there is a lot of subjectivity in essay grading. For example, one grader may take nothing off for a wrong conclusion if the analysis is solid while a different grader may take points off. The November 2008 issue of the Bar Examiner stated that essay questions were weak assessment tools: (1) in part because of the inherent limits on sampling; and (2) it is likely impossible to even get score agreement between raters. The process of grading essays is just too complex - one rater may be angered by illegible writing, another by deficient grammar or spelling, another by poor sentence structure, and a fourth by poor arguments and inadequate knowledge. According to the article, the reliability of the MBE scaled score is 0.90. NCBE found that for the essays to have a reliability of 0.90, they needed to be 16 hours long with 32 different essay questions. Thus, you can’t expect reliability in your essay scores when all you answer are 5-6 essays. The current reliability of the MBE is .92 which means 85% of your current MBE score can be explained by your past MBE score. Thus, if you scored 65% or better in MBE practice on representative questions, you stand a great chance at passing.

oldmanroyal

New
Posts: 1
Joined: Sat Apr 06, 2019 6:49 pm

Re: The "Ask @JoeSeperac" Thread

Postby oldmanroyal » Sat Apr 06, 2019 6:53 pm

Been 14 years since I past the bar, almost 5 years since I practiced. I'm looking for best advice to get the MBE subjects back in my brain. Any suggestions for relearning (basically from scratch) the material?

Smiddywesson

Bronze
Posts: 164
Joined: Tue Feb 12, 2019 1:11 pm

Re: The "Ask @JoeSeperac" Thread

Postby Smiddywesson » Mon Apr 08, 2019 10:46 am

oldmanroyal wrote:Been 14 years since I past the bar, almost 5 years since I practiced. I'm looking for best advice to get the MBE subjects back in my brain. Any suggestions for relearning (basically from scratch) the material?


Yes, I was out of the law for 30 years. It's easy to get back to this if you know where to begin. If you want a program, I'd recommend JoeSeparac. If you are going it on your own, start with the four OPE Online Exams published by the NCBE. They should reflect the material in the subject matter outline. After that, I'd use a combination of Adaptibar and Barbri questions.

When using Adaptibar, I made four Word documents and cut and paste any rules or tips I derived from reading the answers. Read the answers. If you read all of them, they will correct for what are occasionally weak explanations. I reviewed these seven outlines, deleting items as I mastered them, thereby creating condensed outlines as exam day approached.

Initially, you are going to have to push to stay within time limits. I read the short fact patterns slow, and the long ones fast. There are a LOT of simple fact patterns with red herrings that waste a boatload of time. Burn through those bad boys. It's counterintuitive, but you have to do it. I also read the call once, after the fact pattern. A lot of people tell you to read it first, but I kept records, and that didn't help all that often, certainly not enough to counteract the problem of reading it first increases the number of dumb mistakes. You can't read it twice at first, even if it only takes ten seconds, 10 x 200 questions equals 33 minutes in a six hour test.
Finally, your time will greatly improve, thereby allowing you to read the call twice, and all of the answers, even when you think you know which one is correct. That also increases your score.

Finally, understand this is a frustrating process. You WILL, for example, get worse at CivPro as you add Property, and continue to backslide every time you add another subject. Most importantly, understand that this is not a linear process, so you won't see steady improvement until the end. One day you will be a hero, and the next day you will be a dummy. That's normal too.
Protect your psyche. When I was tired or hung over, I did more research and study and less questions. There's so much that goes into performance, you don't want to beat yourself up over the obvious. You will also hit barriers along the way. I hit a plateau at 68%, 75%, and 85%. I finished in the low 90s by changing my time management and instituting the aforementioned methods. There's really no limit on how well you do, the real issue is you just run out of time.

Use the analytics, they will help you sleep at night when you see how wretched everyone is, and steer you in the right direction with studying subtopics.

Good luck.

Anon-non-anon

New
Posts: 11
Joined: Fri Mar 15, 2019 12:40 pm

Re: The "Ask @JoeSeperac" Thread

Postby Anon-non-anon » Mon Apr 08, 2019 11:30 am

JoeSeperac wrote:
Anon-non-anon wrote:Hi Joe,

Way late but just came across this thread and now I'm very curious. I took the July 2016 bar in NY and got a total NY bar score of 341 and scaled MBE of 175.2. Where does that put me?

Thanks!

Sorry for the delayed follow up. Based on your scaled MBE score of 175.2, your estimated raw MBE score was about 164/175 correct (based on my estimate of the J16 MBE scale). This means you answered about 93.7% of the graded MBE questions correctly. This places you in the 99.1% percentile for the MBE. This means that 0.9% of Jul examinees nationwide did better than you on the MBE based on your scaled MBE score of 175.2 (based on July national data for the past 7 years).

Based on a total score of 341, this means your written score was 165.8, which would have placed you in the 93.8% percentile among examinees nationwide (meaning that 6.2% of examinees nationwide would have scored better than you on the MEE/MPT).

If you don't mind (and can still remember), how many questions did you answer in practice, from what sources (e.g. Barbri, Kaplan, Adaptibar, NCBE) and what was your overall % correct?

Joe



Hi Joe,

Thanks so much and of course I don't mind! Least I can do is share what I remember (which is limited). My general approach was to focus on multiple choice, as I had read you can do really badly on the essays and still pass. Plus you can always BS an essay.

I used Kaplan, and focused almost exclusively on doing multiple choice questions once I got through the video modules on each topic. I didn't do any of the review "outlining" (they program in like an hour to outline an entire topic, it was totally non-doable). I did/outlined a couple essays/week, but that was really it other than doing multiple choice questions and limited review of the "handout" stuff you fill in with the videos.

On multiple choice, I used the big practice book and the phone app (which was buggy but convenient) with all topics. I would do between 10 and 50 and then check my answers. No idea how many I did in total, but I would spend about 4-6 hours each day doing them. By the end I was getting around 90%.

I thought Kaplan did a great job with the multiple choice questions. All my friends that did Barbri said they didn't feel prepared. For me, by the end, the basic structure was familiar for all of them. I just spent the summer hanging out on my roof filling in bubbles or doing quizzes on my phone. It was actually a pretty good time.

User avatar
mysojuli

New
Posts: 58
Joined: Mon Jun 09, 2014 11:20 pm

Re: The "Ask @JoeSeperac" Thread

Postby mysojuli » Mon Apr 08, 2019 8:11 pm

Smiddywesson wrote:
oldmanroyal wrote:Been 14 years since I past the bar, almost 5 years since I practiced. I'm looking for best advice to get the MBE subjects back in my brain. Any suggestions for relearning (basically from scratch) the material?


Yes, I was out of the law for 30 years. It's easy to get back to this if you know where to begin. If you want a program, I'd recommend JoeSeparac. If you are going it on your own, start with the four OPE Online Exams published by the NCBE. They should reflect the material in the subject matter outline. After that, I'd use a combination of Adaptibar and Barbri questions.

When using Adaptibar, I made four Word documents and cut and paste any rules or tips I derived from reading the answers. Read the answers. If you read all of them, they will correct for what are occasionally weak explanations. I reviewed these seven outlines, deleting items as I mastered them, thereby creating condensed outlines as exam day approached.

Initially, you are going to have to push to stay within time limits. I read the short fact patterns slow, and the long ones fast. There are a LOT of simple fact patterns with red herrings that waste a boatload of time. Burn through those bad boys. It's counterintuitive, but you have to do it. I also read the call once, after the fact pattern. A lot of people tell you to read it first, but I kept records, and that didn't help all that often, certainly not enough to counteract the problem of reading it first increases the number of dumb mistakes. You can't read it twice at first, even if it only takes ten seconds, 10 x 200 questions equals 33 minutes in a six hour test.
Finally, your time will greatly improve, thereby allowing you to read the call twice, and all of the answers, even when you think you know which one is correct. That also increases your score.

Finally, understand this is a frustrating process. You WILL, for example, get worse at CivPro as you add Property, and continue to backslide every time you add another subject. Most importantly, understand that this is not a linear process, so you won't see steady improvement until the end. One day you will be a hero, and the next day you will be a dummy. That's normal too.
Protect your psyche. When I was tired or hung over, I did more research and study and less questions. There's so much that goes into performance, you don't want to beat yourself up over the obvious. You will also hit barriers along the way. I hit a plateau at 68%, 75%, and 85%. I finished in the low 90s by changing my time management and instituting the aforementioned methods. There's really no limit on how well you do, the real issue is you just run out of time.

Use the analytics, they will help you sleep at night when you see how wretched everyone is, and steer you in the right direction with studying subtopics.

Good luck.


Smiddy you've been constantly giving some GREAT advice. I've Screenshotted all of your tips/tricks and advice. When you pass (or maybe now??) you should make one master post condensing all your advice into one post. I think it would be full of great insights!

Smiddywesson

Bronze
Posts: 164
Joined: Tue Feb 12, 2019 1:11 pm

Re: The "Ask @JoeSeperac" Thread

Postby Smiddywesson » Tue Apr 09, 2019 10:28 am

mysojuli wrote:
Smiddywesson wrote:
oldmanroyal wrote:Been 14 years since I past the bar, almost 5 years since I practiced. I'm looking for best advice to get the MBE subjects back in my brain. Any suggestions for relearning (basically from scratch) the material?


Yes, I was out of the law for 30 years. It's easy to get back to this if you know where to begin. If you want a program, I'd recommend JoeSeparac. If you are going it on your own, start with the four OPE Online Exams published by the NCBE. They should reflect the material in the subject matter outline. After that, I'd use a combination of Adaptibar and Barbri questions.

When using Adaptibar, I made four Word documents and cut and paste any rules or tips I derived from reading the answers. Read the answers. If you read all of them, they will correct for what are occasionally weak explanations. I reviewed these seven outlines, deleting items as I mastered them, thereby creating condensed outlines as exam day approached.

Initially, you are going to have to push to stay within time limits. I read the short fact patterns slow, and the long ones fast. There are a LOT of simple fact patterns with red herrings that waste a boatload of time. Burn through those bad boys. It's counterintuitive, but you have to do it. I also read the call once, after the fact pattern. A lot of people tell you to read it first, but I kept records, and that didn't help all that often, certainly not enough to counteract the problem of reading it first increases the number of dumb mistakes. You can't read it twice at first, even if it only takes ten seconds, 10 x 200 questions equals 33 minutes in a six hour test.
Finally, your time will greatly improve, thereby allowing you to read the call twice, and all of the answers, even when you think you know which one is correct. That also increases your score.

Finally, understand this is a frustrating process. You WILL, for example, get worse at CivPro as you add Property, and continue to backslide every time you add another subject. Most importantly, understand that this is not a linear process, so you won't see steady improvement until the end. One day you will be a hero, and the next day you will be a dummy. That's normal too.
Protect your psyche. When I was tired or hung over, I did more research and study and less questions. There's so much that goes into performance, you don't want to beat yourself up over the obvious. You will also hit barriers along the way. I hit a plateau at 68%, 75%, and 85%. I finished in the low 90s by changing my time management and instituting the aforementioned methods. There's really no limit on how well you do, the real issue is you just run out of time.

Use the analytics, they will help you sleep at night when you see how wretched everyone is, and steer you in the right direction with studying subtopics.

Good luck.


Smiddy you've been constantly giving some GREAT advice. I've Screenshotted all of your tips/tricks and advice. When you pass (or maybe now??) you should make one master post condensing all your advice into one post. I think it would be full of great insights!


Well that was very nice. Thanks. I certainly have a lot of notes, however if I published some of my opinions on why the NCBE keeps ramping up security procedures, and why certain bar companies write their own questions, I'm fairly sure I'd get my ass sued. :lol:



Return to “Bar Exam Prep and Discussion Forum?

Who is online

Users browsing this forum: LawSchoolLA23, Majestic-12 [Bot], RoddyAA and 34 guests