Seperac UBE Score Estimator
Using data from past NY BOLE and NCBE studies, the following calculator will estimate UBE bar exam scores based on the demographic and grade information you enter. Please note that the MBE Practice Information is not required (it has no bearing on your estimated score) but I intend to incorporate this information into future iterations of the calculator. Accordingly, if you are submitting this form after the completion of your bar preparation, this information is greatly appreciated to help improve the accuracy of the calculator. It is important that you read the following explanation of the UBE Score Estimator because while the calculator estimates very well for some demographics, it does a poor job for other demographics where little data is available. Please note that if you would like to receive a detailed breakdown of these statistics/correlations (including score ranges based on standard deviations), please also enter your email below and then press Submit:
The UBE Score Estimator is fairly new although this is the 3rd iteration of the calculator. So far, about 2,000 examinees have sent me their pre-exam stats (including their emails), but only about 150 have followed up with their post-exam results (so please follow up to help). Therefore, keep in mind the calculator adjustments are based on a very small sample. After every exam, I follow up with the most recent batch and then update the calculator again. I also track exam scores - to date, I've received scores from about 5,500 failing examinees (I provide free score reports to failing examinees as a quid pro quo). Recent enhancements to this UBE Score Estimator include adjustments based on law school tier (for domestic examinees) and country of education for Foreign examinees. If you have taken other bar exams, but are taking a certain exam for the first time, you should put "1" for number of bar attempts since bar examiners generally base their statistics on an examinee’s first attempt at that specific exam. The MBE and UBE score estimates are reasonably accurate – for the estimated scaled MBE score, the calculator's maximum deviation was +12/-9 with an average deviation of -0.4 while for the estimated total UBE score, the calculator's maximum deviation was +28/-23 with an average deviation of -2.6 based on responses to date. These statistics will either make you feel more confident or remind you that more work needs to be put into the exam. For more detailed information on these statistics/correlations (including score ranges
based on standard deviations), you should enter your email and
Submit the form (I will then email you back in a few days).
The calculator does a good job of estimating the scores of First Time Domestic examinees because there a lot of solid data available. The calculator is much more unreliable with repeaters and foreign examinees because there is much less data for these demographics. As I collect more data from examinees, the calculator should get more reliable, but currently only 2%-4% of examinees who use the calculator follow up with me with their results so it will be a while. If the calculator estimates you to pass by 20+ points, you are likely going to pass assuming you are doing all the things typical of a first-time examinee which essentially consists of studying full time for about two months and taking a reputable bar review course (e.g. Barbri/Kaplan/Themis). If the calculator estimates you to pass by 20+ points and you are at 60-65% correct or better in overall MBE practice, I would expect you to pass comfortably. Basically, take the UBE Score estimate with a grain of salt, but look at your current MBE practice scores as the final arbiter. For example, if the calculator has you passing by 21 points and you are at 70% correct or better overall in MBE practice, you are in very good territory and the results will likely be on point. However, if you are only 60% correct overall in MBE practice, you need to focus on the exam more (particularly the MBE) because your results will likely shade down (making it possibly too close to call).
However, please note that even with the recent enhancements, the estimates are less accurate (and possibly misleading) for foreign examinees, lower tier examinees, and multiple re-takers because there is less released data for these demographics. The calculator determines the mean total score for a particular demographic and then adjusts based on other criteria such as MPRE/UGPA/LSAT/LPGA. For example, for the MPRE, I calculate the examinee's % correct on the MPRE and then estimate an MBE score based on that. I then attribute a weight. For example, according to NCBE, there is a moderately high relationship between MPRE scores and MBE scores (correlation of .58) and NCBE also states that MBE scores are a surrogate for total bar exam scores since MBE scores are highly related to total bar exam scores. Since there is much more data for First-Time takers as compared to Repeaters, the calculator estimates the scores of First-Time takers most accurately. For foreign examinees, the only adjustment that can be made is with the MPRE and Foreign Country (Canada, Australia, China, etc). However, foreign examinee generally have a pass rate of 50% or less on the bar exam (meaning these examinees are more likely to fail the exam than pass it). Since foreign examinee demographic means are usually below passing, with the absense of other adjustments, the calculator often has the average foreign examinee failing the exam. For multiple re-takers, there is simply not a lot of data from NCBE or the state bar examiners. For lower tier examinees, very few follow up with me after the exam so it makes it difficult to assess the impact of the average scores on lower tier examinees. Finally, extreme inputs will cause problems. While the estimator does a good job of estimating average scores, it does not do well estimating extreme scores. For example, if an examinee has an LGPA of 4.0, this indicates an MBE of 170 based on the study I use for LGPA, but if the examinee is from a T-4 school, the examinee's T-4 status negatively affects the estimate. However, I can’t assess how negatively the estimate is affected because very few T4 examinees follow up with me after the exam to provide data on this. Once I collect more data, I expect future iterations of this calculator to be more accurate. However, as discussed above, your MBE practice scores, assuming the MBE practice questions are of sufficient difficulty and representative of the topics tested, will give you the most insight as to whether or not you will pass the UBE. I hope after a few more iterations, the calculator will do a better job with predicting outcomes for foreign examinees, but even looking at First Time Foreign examinees in July, they are still generally more likely to fail than to pass (which means the projected score will be below 266 rather than above it). However, at the moment, there is a huge void in data regarding these demographics so keep in mind that the predictions are simply estimates and by no means are they always indicative of the actual outcome. As you will see from the additional information (if you submit your info and email to me), at two standard deviations (95% of the demographic), the scores range significantly. I hope to collect enough data to be more accurate at these demographics, but it honestly will be a while for that.
The calculator will predict a "PASS", "FAIL" or "TOO CLOSE TO CALL" based on the expected total score for the average examinee in that demographic. Please keep in mind that these statistics are merely estimates (as you can see from the two standard deviation statistics that I email you if you submit your email, scores can vary widely, meaning that even if the average examinee in that demographic is predicted to fail, an above-average examinee in that demographic may not). Your MBE practice scores, assuming the MBE practice questions are of sufficient difficulty and representative of the topics tested, will give you the most insight as to whether or not you will pass the UBE. The wide range of scores for a particular demographic is likely due to the amount of time an examinee puts into studying for the exam and their ability to score well on the MBE. Doing well on the MBE involves a combination of knowledge and test-taking skills (and skills require drills). Unless you have a solid base for both (e.g. recent law school graduate and/or good test taker based on LSAT, MPRE, LPGA), developing this knowledge and skill takes a lot of time, especially for lower ability examinees – thus if you don’t have a lot of time to spend studying/practicing for the MBE, it is hard to do well on it. While study for the other components of the exam can be “abbreviated” to some extent, MBE study/practice really can’t be given short-shrift, nor can MBE answers be “bluffed” as with the MEE/MPT.
The majority of studies I rely on are based on NY examinees, so any other state will be less accurate, but it still should give you a good idea of where you stand. For example, I can estimate your MBE score from your UGPA and MPRE scores using NCBE studies. I can estimate your pass rate based on your LSAT. I essentially take a number of different points and then arrive at an estimation. Accordingly, if you try to use this calculator for a non-UBE state, you must keep in mind that it is primarily based on NYBOLE and NCBE studies and then I adjust it as I collect data from examinees. However, I find that the demographic estimates usually carry over to other states so you can probably receive a fair estimate of your odds of passing the state (e.g. California). For example, if you choose California from the choice list, you will see 288 as the passing score. This is the 1440/2000 CA passing score converted to the UBE 400 scale. So if you receive an estimate of 20 points or higher than 288 (e.g. 320), there is statistically a good chance you will pass California. I try to take all the demographic estimates to come up with a final estimate. It is sometimes on point and sometimes it is not. A big problem is that very few examinees (less than 3%) who submit their info follow up with me after the exam to let me know the accuracy (I send a single email at the end of the year to ask), so it could be confirmation bias where only the passers who were on-point respond back.
Following are a few comments from examinees who used the estimator and later followed up with me:
Hello! I just got my results and want to provide you with my numbers so you can continue to provide this amazing service. It really helped me when I was super anxious.
I am happy to report I passed the Ohio bar exam with a score of 315. I attached my score report to this email. Before my results came out, I used your UBE Score Estimator (with North Carolina since Ohio wasn’t an option–NC also requires 270 to pass). The Score Estimator predicted 305 for me based on my data, see below. Here is my quick testimony: I’m not sure why but I was convinced I failed. I cried the entire night after the exam was over, couldn’t even fall asleep because I couldn’t stop crying. Although I used Barbri’s program for MBE, MEE, and MPT, I felt like the actual MBE questions were worded differently from Barbri’s questions, and I felt confident on less than 20 of them. Same with the essays; they seemed worded differently and were longer and wordier than Barbri’s practice MEE essays. I also suffered a couple software blackouts during the exam, which increased my anxiety about it. So I was so relieved to see my passing score! My takeaway is that how you feel after the exam does not necessarily correlate to your performance. Let me know if you have any questions. I am happy to contribute data to your business; your score estimator helped me when I was feeling very bad after the exam.
The predictive power of your research is uncanny. I scored a 338 on the October Remote exam. Happy thanksgiving!
My score estimate didn't really affect my studying except to the extent that it seriously reduced my bar-related anxiety, which probably made me more productive and focused. I continued studying the same way and in the same amount, but with more confidence. (I really appreciated the anxiety-reduction tool, though, so thanks so much for working so hard on it!)”
I received my ube score report today and the ube score was on the dot to what your calculator predicted! That's insane! The mbe was off a bit, but still! ... The calculator did ease my fears a good bit. I'm still a worrier by nature so I still worried, but it really did help. It allowed me to think, in the back of my mind, that I'd be ok.
While I'm not sure it had an impact on my studying, I certainly slept better a couple of nights after receiving your estimator. In fact, I looked at it on several occasions when I had any feelings of self doubt. That alone was worth it to me. I do hope students in similar situations have the luxury of receiving your estimator and enjoy at least a couple of nights stress free.
I think the score estimate was helpful in calming me and giving me a sense of confidence.
My estimated score helped motivate me to stay on course for the bar study program and to worry less about failing as long as I finished the program. It very much helped to calm my nerves because I felt that I would fail until I did your score calculator and found out that I was already in a good place before I started.
I wanted to let you know you were pretty close in your analysis! I scored a 290 on the WA Bar exam. My MBE was 134.7 which was lower than expected, but my written was 155.6 - higher than expected. Your calculator was a factor that gave me confidence to go in believing I could pass with hard work.
Just wanted to send you an email comparing your prediction to my actual score. It gave me solace to use the calculator after the exam. As such, I wanted to give feedback to help your analysis for others.
The estimator and calculators were massively helpful in keeping me sane over the summer and while waiting for results.
The free stuff on your website, including the score estimator, were really helpful and comforting to me while I was preparing for the exam.
Thanks again for this breakdown. It really helped keep me motivated when preparing for the exam. I wanted to follow up now, because... I PASSED!! I got a 141.4 MBE, and a 272 UBE overall. So, I'm actually a bit disgusted at how close your analysis was -- pretty hilarious. Hope you can use these data points in refining your software, and help others in the future in the same way you helped me!
your calculator made me feel a lot better while I was waiting for the results ;)
Thank you again! Had I not taken your score estimator, I would not have known what I was capable of doing. It helped me keep pushing during my second crack at the exam.
I wanted to write you with an update on my scores. I took the Illinois Bar in Feb 2020. I passed with a total score of 301 (enough to waive in to every UBE jurisdiction), making your prediction of a score of 298 pretty spot on. Thanks again for reducing my stress a little by providing your predictions.
The results should automatically update when you enter/change a selection. Please note that as a fairly new calculator, there may be some calculation errors, so if you see a result that appears incorrect, please email me at joe@seperac.com.
Please keep in mind that while this Calculator provides what I regard as a fairly reliable estimate, it is still just an estimation. For example, if your demographic is July exam: Domestic-Educated: All Takers: Male and Female: Number of Bar Attempts: 1, the average scaled MBE score was 145.3 (which is about 68% correct) and the average total score was 291 (converted from pre-UBE exam scores). So the average Domestic Educated First-Time taker is going to score well above passing. However, if you submit your form (along with your email so I can follow up with you post-exam), I provide a better breakdown of the stats. For example, in this case, 95% of examinees in this demographic had a scaled MBE score between 115.8-174.9 and a total UBE score between 240-342. So then I adjust scores based on other studies, but this is obviously an inexact science. For example, based on a 2007 NYBOLE study, someone with an LSAT score of 180 had a 99.5% probability of passing. So if you put in 180 for LSAT, the calculator will shade your score up, but not dramatically. This is because I have seen examinees with 170+ LSATs fail. However, if all your stats are good (e.g. T-1, high UPGA, LGPA, MPRE, etc), the estimated score of the calculator will arrive at 340 or so (well above the original assessment of 291). With each iteration, it will get more precise as I go through the actual results and see which variable has the greatest impact on score. My advice is to take this calculator with a grain of salt, but look at your current MBE practice scores as the final arbiter. At some point I will probably add current MBE practice scores as a variable in the Estimator, but I want to collect a lot of data first. Accordingly, these calculator results are based on the average scores for the demographics you entered, and you can be well above or below this average.
Please note that if you would like to receive a detailed breakdown of these statistics/correlations (including score ranges based on standard deviations), please also enter your email below and then press Submit. Following is an example of this extra information that I email to you, assuming you provide me with your email. If you submit your information, I will then follow up with you at the end of December to check on the accuracy of the estimation:
Thanks for submitting your UBE information. Following are more detailed statistics/correlations on your projected performance of 172 on the MBE and 341 on the UBE:
NYBOLE released studies of the July 2005 and Feb 2006 exams which examined the impact of the increase in the passing score on the New York Bar Examination, and the following demographic profile statistics are applicable to you. See http://www.nybarexam.org/press/ncberep.pdf and http://www.nybarexam.org/press/nyrep_feb06.pdf
Based on your demographic of July exam: Domestic-Educated: All Takers: Male and Female: Number of Bar Attempts: 1, the average scaled MBE score was 145.3 (which is about 68% correct) and the average total score was 291 (converted from pre-UBE exam scores). Looking further, 95% of examinees in your demographic (2 SDs) had a scaled MBE score between 115.8-174.9 and a total UBE score between 240-342.
Based on your demographic of July exam: Domestic-Educated: First-Time Takers: Male: Sex: Male, the average scaled MBE score was 148 (which is about 70% correct) and the average total score was 292 (converted from pre-UBE exam scores). Looking further, 95% of examinees in your demographic (2 SDs) had a scaled MBE score between 118.8-177.2 and a total UBE score between 240-344.
Based on your demographic of July exam: Domestic-Educated: First-Time Takers: Male and Female: Racial/Ethnic Groups: Caucasian/White, the average scaled MBE score was 147.1 (which is about 69% correct) and the average total score was 294 (converted from pre-UBE exam scores). Looking further, 95% of examinees in your demographic (2 SDs) had a scaled MBE score between 118.4-175.8 and a total UBE score between 245-344.
According to NCBE's December 2011 Bar Examiner, students with higher UGPAs tend to have higher MBE scores. According to NCBE, the correlation between UGPA and the MBE is 0.36 (a moderate relationship); the correlation between LSAT and the MBE is 0.57 (a moderately high relationship); and the correlation between LGPA and the MBE is between 0.55-0.70 depending on the law school (a moderate to high relationship).
Based on your MPRE score of 127, you answered about 85% correct on the MPRE. According to NCBE, there is a moderately high relationship between MPRE scores and MBE scores (correlation of .58). NCBE also states that MBE scores are a surrogate for total bar exam scores since MBE scores are highly related to total bar exam scores. see http://www.ncbex.org/assets/media_files/Bar-Examiner/articles/2011/800411Testing.pdf. If you likewise answered 85% correct on the MBE, you should expect to see a scaled MBE score of about 164 (this is based on the 2013 raw to scaled MBE conversion in New York).
Based on your UGPA of 3.8, your expected MBE score is about 155. See http://www.ncbex.org/assets/media_files/Bar-Examiner/articles/2011/800411Testing.pdf
Based on your LSAT score of 166, you did better than 93% of examinees who took the LSAT. See http://www.alphascore.com/resources/lsat-score-conversion/
Based on a 2007 study entitled 'Unpacking The Bar-Of Cut Scores And Competence' by Gary S. Rosin, with your LSAT score of 166, your expected bar passage rate is 95%. See https://papers.ssrn.com/sol3/papers.cfm?abstract_id=988429.
Examinees who perform better on the LSAT generally perform better on the MBE than those who show a poor performance on the LSAT. According to NCBE, based on your LSAT of 166, your estimated scaled MBE score is about 157. See http://www.ncbex.org/assets/media_files/Bar-Examiner/articles/2011/800411Testing.pdf
Based on a 2007 NYBOLE study entitled 'New York Bar Examination Performance in February and July 2006 for Candidates Failing for the First Time in July 2005', with your LSAT score of 166, the probability of passing for a domestic educated first time taker is 94.6%. See http://www.nybarexam.org/press/ncberep2.pdf.
Based on a 2007 NYBOLE study entitled 'New York Bar Examination Performance in February and July 2006 for Candidates Failing for the First Time in July 2005', with your LGPA of 3.8, the probability of passing for a domestic educated first time taker is 99.2%. See http://www.nybarexam.org/press/ncberep2.pdf.
For submitting your demographic information, following is a $25 coupon code to any one of my February 2020 subscription modules available at www.mberules.com or www.ubeessays.com. Please note that this discount code expires February 15, 2020:
If you take the exam and think you may not have passed (especially if you are an at-risk demographic), I suggest you complete my Post Exam form.
Filling out this form immediately after you take the exam (while the information is still fresh in your mind) can help you later. For example, using this information, I track the key details of your attempt, so if you later find that you failed the exam, I will try to match your responses/statistics to whoever previously submitted the most comparable details (and later passed) to give you their advice on what worked for them.