The Testing Column: July 2019 MBE: Here Comes the Sun; August 2019 MPRE: Here Comes the Computer

This article originally appeared in The Bar Examiner print edition, Fall 2019 (Vol. 88, No. 3), pp 33–35.

By Mark A. Albanese, PhD

A Rebound in the July 2019 MBE Score

In July 2019, the mean score on the Multistate Bar Examination (MBE) was 141.1, a 1.6-point increase from the mean score in July 2018 and the largest increase compared to the previous July’s mean since the increase seen in July 2008. Of 52 jurisdictions (testing 10 or more examinees), 47 had higher MBE scores in July 2019 than a year earlier. Based upon MBE scores alone (jurisdictions include scaled written scores in determining their final pass rates), the estimated national pass rate is projected to be 63%, up roughly 3 percentage points from July 2018.

This follows the 1.2-point increase in the February 2019 MBE mean score over that of February 2018—an increase that I indicated in my previous column might be the light at the end of the tunnel following the historic low MBE means in February and July 2018. I was afraid the light might actually be an oncoming train, so the July 2019 results were a definite relief.

The surprise was not quite as great as in February because, unlike in February, this time I made use of prediction equations I mentioned in my last column, and all seven of the different equations I employed predicted an increase in the MBE mean. Depending on the equation, the rise was predicted to be somewhere between 0.2 and 2.0 points, with five of them being within a half point of the actual 1.6-point value.

The only problem is that we statisticians also recognize that statistics come with a margin for error—which, in the case of relatively small changes found in the MBE mean, covers the range from catastrophic declines to historic rises. That gives us pause in making any specific predictions about the future.

The following are some additional characteristics of the July 2019 bar exam:

• Number of examinees: The July 2019 exam was taken by 45,334 examinees, only 60 more than in 2018. Except for July 2018, this is still the lowest number taking the July bar exam since 2001.
• Reliability: The reliability for the July 2019 exam was 0.94, the highest in MBE history, indicating that scores were the most accurately estimated in MBE history.
• Likely first-time takers: The examinees likely to be taking the bar exam for the first time constituted about 65%, which was almost identical to July 2018; the likely first-time taker mean was up 2.2 points from last year, which at 145.1 was the highest first-time taker value since 2013.
• Likely repeaters: The mean of examinees likely repeating the bar exam was 131.3, an increase of 0.2 points over the mean of those likely repeating in 2018.
• LSAT scores: First-time takers in July 2019 would largely be examinees who entered law school in 2016. The mean LSAT score at the 25th percentile for the 2016 law school entrants was 152.55, an increase of 0.18 points from that of the 2015 law school entrants who would likely have taken the bar exam in July 2018.
• LSATs administered: The number of LSATs administered was also up for the 2016 law school entering class compared to the 2015 entering class.1 The number of LSATs administered tends to coincide with the number of law school applicants. As the number of applicants to law school goes up, law schools can be more selective in the students they admit, which generally leads to law school classes having better credentials (higher LSAT scores and entering GPAs), which tends to lead to higher performance in law school and, ultimately, on the bar examination.

The 2016 increase in the number of LSATs administered was the first increase since the 2010 entering class. This upward trend in the number of LSATs administered continued in 2017. The number of LSATs administered to the 2017 entering class, which will compose the bulk of those taking the bar exam in July 2020, was up 3.3%, while the number of LSATs administered to those entering law school in 2018 was up a whopping 18.1%, reaching levels not seen since the 2012 entering class. This bodes particularly well for the July 2021 bar exam, when these students will be on track to graduate. The 2019 entering class, who will be taking the July 2022 bar exam in large numbers, saw another 7.3% increase in LSATs administered over the previous year, reaching levels not seen since the 2011 entering class.2

So, we have seen the MBE light and it appears to be the sun, not an oncoming train.

MPRE Computer-Based Administration

In August, over 4,000 candidates took the Multistate Professional Responsibility Examination (MPRE) at Pearson VUE centers across the country for the inaugural computer-­based testing (CBT) administration of the MPRE. The remaining nearly 13,000 candidates took the paper-based exam at testing centers administered by the Law School Admission Council.

A phased approach for this type of transition to CBT is recommended by experts in testing to help ensure that the transition happens smoothly. Our main goal with the August administration was to confirm that scores obtained via CBT are consistent with those obtained via paper-based administration. Indeed, we found no meaningful differences between the scores obtained by the two groups. In November, a similar number of candidates will take the computer-­based MPRE; the purpose of the November administration is to solidify testing procedures for a computer-based environment, including testing conditions for candidates who have been granted accommodations for the MPRE. These two smaller computer-based administrations will allow us to minimize any problems that may occur upon the full transition to CBT in March 2020.

Beginning with the March 2020 administration, the major difference for all candidates in terms of registration will be that instead of registering to take the MPRE in a single two-hour testing period at a site administered by the Law School Admission Council, candidates will need to schedule a two-hour test appointment at a Pearson VUE center on one of two possible test days for a given administration. Since each center has limited capacity, candidates should schedule their testing appointments as early as possible to maximize the dates, times, and locations available to them. Seats will be provided on a first-come, first-served basis, so late registrants may have to travel to a center that is not as close as one they might prefer. Also, a new MPRE test accommodations process has been implemented for 2020; details are available at www.ncbex.org/exams/mpre/ada-accommodations/.

As added incentive for the March 2020 administration, if candidates have any issues with taking a high-stakes test on Friday the 13th, early registration for the Thursday, March 12, date might be in their best interest.

Contact us to request a pdf file of the original article as it appeared in the print edition.