Bar grading? Forum

[ConfusedL1]

Can anyone explain how grading actually works on the bar for UBE jurisdictions? So each state can use its own point system? Are the points for every essay on the MEE added together as a raw score for the 25% and then weighted?

[JDAdvisingLLC]

Each jurisdiction has their own scale for the MEE and MPTs -- (Some use 1-6; other use 1-10, New York uses 20-50 to grade the MEEs.)

The MBE is scored nationally so it is always on a 200-point scale and is the same everywhere. (They only tell you your "converted" or scaled MBE score rather than your raw score).

For UBE jurisdictions:
The MEE's are always weighted at 30%.
The MPTs are always weighted at 20%
The MBE is weighted at 50%.

After the scores are scaled to meet those above requirements you will get an overall score of, say, 260 or 280 or whatever and that is what you use to compare yourself to passing scores in other jurisdictions.

So the weighting is the same but the actual points that you see for each essay will vary depending on the jurisdiction you are in.

I'm not sure if this is what you were asking, so sorry if it was not!

[JoeSeperac]

New York uses 20-50 to grade the MEEs.

I think this is a typo. According to the F17 NYBOLE score report "The scaled score for each of the six MEE questions and two MPT questions are arrived at by converting the raw score for each question to a scale that generally ranges from approximately 20 to 80, with 50 as the mean."

The best way to understand how the scores are calculated is to see it in action. My July 2016 NY UBE calculator should estimate the July 2017 NY UBE exam fairly closely:
http://seperac.com/zcalc-ube-j16.php

The default values are the averages from the examinees that failed and sent me their scores (meaning if you get a 120 on the MBE, you will probably fail too). Even though NY says that 50 is the mean, it is not the passing score (this changes from exam to exam). For example, for the J16 NY UBE exam, any MEE or MPT score above 47.82 was a passing score. This means if you put 47.82 for every MEE and MPT score into the J16 calculator and then 133 for the MBE, you will have an exactly passing score of 266 calculated for the J16 NY UBE exam.

[JDAdvisingLLC]

Yes that is a typo. I meant 20-80. Thank you for the correction.

[ConfusedL1]

So essays are all graded as one unit? For example on NY if they have six essays that's 480 possible points that is then weighted at 25%?

[JoeSeperac]

So essays are all graded as one unit? For example on NY if they have six essays that's 480 possible points that is then weighted at 25%?

No. Let's go back to the July 2016 NY UBE example. If you received a scaled score of 47.82 on a July 2016 MEE answer, this was an exactly passing score. This would have contributed 13.3 points to your total UBE score (which is 5% of 266). If you had exactly passing scores for all 6 MEE answers, this would have contributed 79.8 points to your total UBE score (which is 30% of 266). In theory, the most you could ever get in points on the MEE is 30% of 400, or 120 points (you would have to write 6 model answers). Interestingly, if you answered none of the 6 MEE essays, you would still receive about 35 points towards your total UBE score (meaning you could pass NY with a 170 MBE, above average MPTs and not a word written for the essays).

If you wrote 2 exactly passing MPTS on the J16 exam (scores of 47.82), the MPT would have contributed 53.2 points to your total UBE score (20% of 266). If you add the 79.8 to the 53.2, you get 133 (an exactly passing written score).

On a side note, if you failed in another UBE state, I plan to make score calculators for those states once I have enough score sheets to figure out the scales. Alternatively, if you have your MBE subscores (e.g CA), I can tell you your raw scores:
http://seperac.com/bar/subscoreform.php

If you have a copy of your essays from your F17 attempt, I will be finishing my F17 MEE/MPT analysis in about a week if you want to participate:

[+] Spoiler

How my MEE/MPT Analysis and Comparison work:
1) If you fail the UBE exam and have a copy of your essays, you email them to me.
2) I transcribe your essays and statistically analyze them.
3) Once I have a large enough sample of essays (usually 20-30 examinees for a February exam or 50-60 examinees for a July exam), I create the MEE/MPT Analysis for that exam.
4) I then email you a free 37+ page essay analysis that statistically compares your MEE/MPT answers to everyone else who sent me their essays. A sample of the J16 Analysis (37 pages) is here:
http://www.seperac.com/pdf/Sample-Essay ... 202016.pdf
This MEE/MPT Analysis is confidential – I don't share your essays with anyone and no one else sees them. All I do is transcribe your essays and make the report once I have a large enough sample. NOTE: I plan to start comparing the MEE essays to the point sheets so the MEE section of the analysis will be similar to the MPT section of the analysis, giving you even more insight into your MEE/MPT answers.
5) When I send you the MEE/MPT Analysis, I ask you if you want to participate in the MEE/MPT Comparison. If you say No, nothing else happens with your essays and no one ever sees them except me. If you say Yes, I include your essays in the MEE/MPT Comparison which lets you see your answers compared to everyone else's side-by-side. Following are very small samples of my February and July 2010 MPT comparisons:
http://www.seperac.com/Feb2010Analysis/
http://www.seperac.com/examinees/JULY2010/
This Comparison is viewable by everyone who participates in it and also by subscribers. In the Comparison, everything is redacted (I even check your MPTs to make sure you didn't mention your name by mistake) so there is nothing to identify you. The majority of examinees decide to participate once they see their Analysis (out of the 500+ examinees that have sent me their essays for the Analysis, only 7 have opted out of the Comparison). I have been doing these Comparisons since 2010 and never has an examinee told me their confidentiality was compromised in any way.
6) If you participate in the Comparison, I give you a $40 coupon code if you decide to later subscribe to the full subscription site (where you can view the Comparisons for other exams) or refund $40 if you are already subscribed.

Both the Analysis and the Comparison are great ways to get some useful insights into your MEEs/MPTs. For example, one of the more useful aspects of the MEE/MPT Analysis is a “Top 20 Words” analysis that reports the top 20 words the above average answers used that you did not. Through this “Top 20 Words” list, I find that failing examinees sometimes fail to use IRAC phrases such as “whether”, “here”, “therefore”, or “however.” This tells me the examinee’s essay is probably not as organized as it could be. Other times the examinees fail to use analysis words such as “because”, “since” , or “as.” In an IRAC analysis, “because” is the single most important word to use when analyzing the facts in the question. The failure to use words such as "because", "since" and "as" will negatively affect the analysis portion of your essay and can only hurt your score. Lastly, examinees often fail to use the legal terminology associated with a particular essay topic. I refer to these terms as “buzz-words.” I believe that graders do not spend a lot of time reviewing an essay, so the failure to use buzz-words (to signal to the grader that you understand the topic) can hurt your essay score. Put simply, absence of such words indicates the examinee may not be discussing things he/she needs to discuss.

The MEE/MPT Comparison is a fantastic way to visually identify what you did wrong on the MEE/MPT by looking at your essays side by side against others. You can look at exactly passing answers to see how much (or how little) is required for an exactly passing score. When you look at the text comparison, you start to see the commonality in language with high scoring answers – in a sense this trains you to include the same language in similar essays. The PDF comparisons (where you view the actual PDFs of the answers side by side) let you see each essay’s layout, structure, and formatting so you can visually learn how to emulate the high scoring answers (and conversely, avoid the styles of the low scoring answers). For example, one examinee (non-subscriber) who failed F15 told me: "I did much better on my essays this time due in large part to your comparison tool. I found that to be extremely helpful." For the F15 exam, this examinee's 5-Essay average was 53.74 (a passing essay average for the Feb 2015 exam was 51.43). Based on 196 submitted score reports, this 5-Essay average was ranked 9/196 (this means the examinee had a 5-Essay average better than 95.4% of the examinees that sent me their Feb 2015 score information). In July 2014, this examinee had a 5-Essay average was 45.22 (a passing essay average for the July 2014 exam was 47.83). Based on 315 submitted score reports, this 5-Essay average was ranked 131/315 (which was better than 58.4% of the examinees that sent me their July 2014 scores).

Thus, with the MEE/MPT Analysis you can statistically compare your MEEs/MPTs to other graded MEEs/MPTs while with the MEE/MPT Comparison, you can visually compare your MEEs/MPTs to other graded MEEs/MPTs. On the subscription site, I have Comparisons for the last 15 MPT exams but only the J16 MEE since NY only switched in J16 (F17 to follow soon). For example, with the July 2015 MPT Comparison (Opinion Letter), you can examine 63 graded MPTs (for a total of 1,800+ comparisons). For the July 2016 MEE Comparison, for each of the six MEE questions, there are 741 comparisons based on 33 examinee essays.

[ConfusedL1]

So essays are all graded as one unit? For example on NY if they have six essays that's 480 possible points that is then weighted at 25%?

No. Let's go back to the July 2016 NY UBE example. If you received a scaled score of 47.82 on a July 2016 MEE answer, this was an exactly passing score. This would have contributed 13.3 points to your total UBE score (which is 5% of 266). If you had exactly passing scores for all 6 MEE answers, this would have contributed 79.8 points to your total UBE score (which is 30% of 266). In theory, the most you could ever get in points on the MEE is 30% of 400, or 120 points (you would have to write 6 model answers). Interestingly, if you answered none of the 6 MEE essays, you would still receive about 35 points towards your total UBE score (meaning you could pass NY with a 170 MBE, above average MPTs and not a word written for the essays).

If you wrote 2 exactly passing MPTS on the J16 exam (scores of 47.82), the MPT would have contributed 53.2 points to your total UBE score (20% of 266). If you add the 79.8 to the 53.2, you get 133 (an exactly passing written score).

On a side note, if you failed in another UBE state, I plan to make score calculators for those states once I have enough score sheets to figure out the scales. Alternatively, if you have your MBE subscores (e.g CA), I can tell you your raw scores:
http://seperac.com/bar/subscoreform.php

If you have a copy of your essays from your F17 attempt, I will be finishing my F17 MEE/MPT analysis in about a week if you want to participate:

[+] Spoiler

How my MEE/MPT Analysis and Comparison work:
1) If you fail the UBE exam and have a copy of your essays, you email them to me.
2) I transcribe your essays and statistically analyze them.
3) Once I have a large enough sample of essays (usually 20-30 examinees for a February exam or 50-60 examinees for a July exam), I create the MEE/MPT Analysis for that exam.
4) I then email you a free 37+ page essay analysis that statistically compares your MEE/MPT answers to everyone else who sent me their essays. A sample of the J16 Analysis (37 pages) is here:
http://www.seperac.com/pdf/Sample-Essay ... 202016.pdf
This MEE/MPT Analysis is confidential – I don't share your essays with anyone and no one else sees them. All I do is transcribe your essays and make the report once I have a large enough sample. NOTE: I plan to start comparing the MEE essays to the point sheets so the MEE section of the analysis will be similar to the MPT section of the analysis, giving you even more insight into your MEE/MPT answers.
5) When I send you the MEE/MPT Analysis, I ask you if you want to participate in the MEE/MPT Comparison. If you say No, nothing else happens with your essays and no one ever sees them except me. If you say Yes, I include your essays in the MEE/MPT Comparison which lets you see your answers compared to everyone else's side-by-side. Following are very small samples of my February and July 2010 MPT comparisons:
http://www.seperac.com/Feb2010Analysis/
http://www.seperac.com/examinees/JULY2010/
This Comparison is viewable by everyone who participates in it and also by subscribers. In the Comparison, everything is redacted (I even check your MPTs to make sure you didn't mention your name by mistake) so there is nothing to identify you. The majority of examinees decide to participate once they see their Analysis (out of the 500+ examinees that have sent me their essays for the Analysis, only 7 have opted out of the Comparison). I have been doing these Comparisons since 2010 and never has an examinee told me their confidentiality was compromised in any way.
6) If you participate in the Comparison, I give you a $40 coupon code if you decide to later subscribe to the full subscription site (where you can view the Comparisons for other exams) or refund $40 if you are already subscribed.

Both the Analysis and the Comparison are great ways to get some useful insights into your MEEs/MPTs. For example, one of the more useful aspects of the MEE/MPT Analysis is a “Top 20 Words” analysis that reports the top 20 words the above average answers used that you did not. Through this “Top 20 Words” list, I find that failing examinees sometimes fail to use IRAC phrases such as “whether”, “here”, “therefore”, or “however.” This tells me the examinee’s essay is probably not as organized as it could be. Other times the examinees fail to use analysis words such as “because”, “since” , or “as.” In an IRAC analysis, “because” is the single most important word to use when analyzing the facts in the question. The failure to use words such as "because", "since" and "as" will negatively affect the analysis portion of your essay and can only hurt your score. Lastly, examinees often fail to use the legal terminology associated with a particular essay topic. I refer to these terms as “buzz-words.” I believe that graders do not spend a lot of time reviewing an essay, so the failure to use buzz-words (to signal to the grader that you understand the topic) can hurt your essay score. Put simply, absence of such words indicates the examinee may not be discussing things he/she needs to discuss.

The MEE/MPT Comparison is a fantastic way to visually identify what you did wrong on the MEE/MPT by looking at your essays side by side against others. You can look at exactly passing answers to see how much (or how little) is required for an exactly passing score. When you look at the text comparison, you start to see the commonality in language with high scoring answers – in a sense this trains you to include the same language in similar essays. The PDF comparisons (where you view the actual PDFs of the answers side by side) let you see each essay’s layout, structure, and formatting so you can visually learn how to emulate the high scoring answers (and conversely, avoid the styles of the low scoring answers). For example, one examinee (non-subscriber) who failed F15 told me: "I did much better on my essays this time due in large part to your comparison tool. I found that to be extremely helpful." For the F15 exam, this examinee's 5-Essay average was 53.74 (a passing essay average for the Feb 2015 exam was 51.43). Based on 196 submitted score reports, this 5-Essay average was ranked 9/196 (this means the examinee had a 5-Essay average better than 95.4% of the examinees that sent me their Feb 2015 score information). In July 2014, this examinee had a 5-Essay average was 45.22 (a passing essay average for the July 2014 exam was 47.83). Based on 315 submitted score reports, this 5-Essay average was ranked 131/315 (which was better than 58.4% of the examinees that sent me their July 2014 scores).

Thus, with the MEE/MPT Analysis you can statistically compare your MEEs/MPTs to other graded MEEs/MPTs while with the MEE/MPT Comparison, you can visually compare your MEEs/MPTs to other graded MEEs/MPTs. On the subscription site, I have Comparisons for the last 15 MPT exams but only the J16 MEE since NY only switched in J16 (F17 to follow soon). For example, with the July 2015 MPT Comparison (Opinion Letter), you can examine 63 graded MPTs (for a total of 1,800+ comparisons). For the July 2016 MEE Comparison, for each of the six MEE questions, there are 741 comparisons based on 33 examinee essays.

Thanks that's pretty helpful. I'm taking the DC bar exam. Things are weighted 50 MBE, 25 MPT, 25 MEE. I believe the essays are scored a 1-6 scale?

I'm not sure how mathematically this works, but i'm trying to figure out what my likelihood of passing would be if I scored 70% correct raw on the MBE and then below passing on both the MPT and MEE. I think that's about the worst I could do.

[JoeSeperac]

Thanks that's pretty helpful. I'm taking the DC bar exam. Things are weighted 50 MBE, 25 MPT, 25 MEE. I believe the essays are scored a 1-6 scale?

I'm not sure how mathematically this works, but i'm trying to figure out what my likelihood of passing would be if I scored 70% correct raw on the MBE and then below passing on both the MPT and MEE. I think that's about the worst I could do.

If you answer 125 questions correctly on the July MBE (71.4% correct based on a total of 175 questions graded), this would result in a scaled MBE score of about 150 on the MBE (based on the 2013 NY bar exam raw-scaled conversion - this is the most recent exam that reported raw/scaled MBE scores). A scaled score of 150 would mean your MBE score was better than about 72% of examinees nationwide for the July MBE. Based on a UBE pass rate of 266, an examinee with a scaled score of 150 on the MBE would need a scaled MEE/MPT score of 116 to pass the exam. If the MEE/MPT percentiles mirror the MBE percentiles (which they should), this means that an examinee with an MBE score of 150 (which is about 71% correct on the MBE) needs to write MEE/MPT answers that are better than only 8% of the July examinees. see http://www.ncbex.org/publications/stati ... tatistics/

As a first-time taker, my UBE Score Estimator will also give you a pretty good estimate of your passing chances:
http://www.seperac.com/zcalc-passcalc.php

[Leprechaun]

JoeSeperac,

I took the Feb bar and had a 156.2 scaled MBE score. What approximate percentage correct is that?

Thanks

[JoeSeperac]

JoeSeperac,

I took the Feb bar and had a 156.2 scaled MBE score. What approximate percentage correct is that?

Thanks

Based on your scaled MBE score of 156.2, your estimated raw MBE score was about 135/175 correct. This is based on the 2013 New York MBE raw/scaled conversion (which is the most recent time an MBE raw/scaled conversion was released). This means you answered about 77% of the graded MBE questions correctly. On the MBE, examinees usually score close to their MBE practice percentage (especially when they have done a large number of questions). Accordingly, I estimate that you likely answered at least 70% correct or better in overall MBE practice (please let me know if this is inaccurate).

Based on the F16 national statistics on the MBE (this year's statistics will not be released until next year), this places you in the 90.4% percentile for the MBE. This means that 9.6% of last year's examinees nationwide did better than you on the MBE based on your scaled MBE score of 156.2. If you let me know your total UBE score, I can tell you how you did nationwide on the MEE/MPT.

[Leprechaun]

JoeSeperac,

I took the Feb bar and had a 156.2 scaled MBE score. What approximate percentage correct is that?

Thanks

Based on your scaled MBE score of 156.2, your estimated raw MBE score was about 135/175 correct. This is based on the 2013 New York MBE raw/scaled conversion (which is the most recent time an MBE raw/scaled conversion was released). This means you answered about 77% of the graded MBE questions correctly. On the MBE, examinees usually score close to their MBE practice percentage (especially when they have done a large number of questions). Accordingly, I estimate that you likely answered at least 70% correct or better in overall MBE practice (please let me know if this is inaccurate).

Based on the F16 national statistics on the MBE (this year's statistics will not be released until next year), this places you in the 90.4% percentile for the MBE. This means that 9.6% of last year's examinees nationwide did better than you on the MBE based on your scaled MBE score of 156.2. If you let me know your total UBE score, I can tell you how you did nationwide on the MEE/MPT.

Thank you for the information. For your data purposes, I took Barbri and got 125 of 200 on the simulated MBE, 35 of 50 on the next major question set, 57 of 100 on the final big MBE practice they had. On the 18 question Barbri sets, I think I averaged 75 to 78% on Torts and Crim Law Procedure, but only averaged 56 to 60% on Property and Evidence, and the other subjects were in between those highs and lows. All in all, I probably worked through about 1100 Barbri questions.

Regarding the UBE, I was in a non UBE state, Texas, and scored a 774 overall, but if you pass, Texas only releases your scaled MBE score and not the MPT, state essays, or procedure & evidence scores, so I don't know how I did on the other sections.

I know to me, the Barbri MBE questions seemed so much harder than the actual MBE. I was pleasantly surprised when I took the real MBE that it seemed much easier.

[JoeSeperac]

Thank you for the information. For your data purposes, I took Barbri and got 125 of 200 on the simulated MBE, 35 of 50 on the next major question set, 57 of 100 on the final big MBE practice they had. On the 18 question Barbri sets, I think I averaged 75 to 78% on Torts and Crim Law Procedure, but only averaged 56 to 60% on Property and Evidence, and the other subjects were in between those highs and lows. All in all, I probably worked through about 1100 Barbri questions.

Regarding the UBE, I was in a non UBE state, Texas, and scored a 774 overall, but if you pass, Texas only releases your scaled MBE score and not the MPT, state essays, or procedure & evidence scores, so I don't know how I did on the other sections.

I know to me, the Barbri MBE questions seemed so much harder than the actual MBE. I was pleasantly surprised when I took the real MBE that it seemed much easier.

So you were at about 67% correct for those 4 subjects using Barbri Qs, which put you in line to score a 150+ on the MBE.

In regards to your BARBRI simulated exam score of 125, a thread on TLS two year's ago resulted in a spreadsheet that compared practice scores to final MBE scores (and final outcomes). The spreadsheet is here:
https://docs.google.com/spreadsheets/d/ ... li=1#gid=0

I took the data from the spreadsheet an sorted it based on the BARBRI PT raw score:
http://seperac.com/pdf/BARBRI%20PT%20Sc ... %20TLS.pdf

Of the 77 examinees who took BARBRI in July 2015 and reported their exact PT scores, 75 passed the exam, including all those with a 113 raw or higher (which is about 57% questions correct).