In my last post, I talked about how Barry and St. Thomas Law Schools “bombed” the Florida Bar in 2016. But in a sense, their graduates didn’t bomb the test at all. In reality, they performed exactly as expected on the bar given their aptitude.
In this column, I want to compare the performance of these schools with what many would consider to be a peer or competitor school in Florida, Nova Southeastern University. Nova Southeastern grads, compared to these two schools, did very well on the 2016 Florida Bar exams, passing at a 64.7% rate. It is important to understand why.
As a starting point let’s compare the entrance credentials of those who were most likely taking the 2016 bar for the first time from these schools -- the full-time entering class of 2013, and the part-time entering class of 2012. (St. Thomas has only a full-time division.)
Admitted Student Profiles from ABA Standard 509 Information Reports
School 75/50/25 LSAT 75/50/25 GPA Class Size (year and division)
Barry 150/147/145 3.21/2.90/2.62 233 (2013 full-time)
Barry 152/147/145 3.38/2.96/2.54 62 (2012 part-time)
St. Thomas 150/147/145 3.33/3.06/2.78 261 (2013 full-time)
Nova SE 152/149/147 3.33/3.09/2.84 243 (2013 full-time)
Nova SE 153/148/145 3.29/3.06/2.76 54 (2012 part-time)
Now let’s see how these students did on the Florida bar in 2016.
2016 Florida Bar Results- Combined February and July – First Time Takers
Takers Passers Percentage
Barry 217 98 45.2%
St. Thomas 180 81 45.0%
Nova SE 221 143 64.7%
As you can see, Barry and St. Thomas had virtually identical entering classes in 2013, and they had virtually identical results on the 2016 bar exam, 3 years later. In contrast, Nova Southeastern, a school of similar size had a nearly 20% higher bar pass rate. Now 64.7% is certainly not great and was still below the state average, but importantly, with a 64.7% first-time pass rate, Nova would almost certainly make the proposed ABA bar passage standard of 75% within two years, whereas Barry and St. Thomas are both likely to fall short with a starting point of 45%.
Now it may be that Nova Southeastern is doing something dramatically different and more effective than St. Thomas and Barry to prepare its students to pass the bar, but it is doubtful that any differences in their approach would account for a 20% difference in bar pass rates among students of similar abilities. The far more likely explanation is simply the difference in LSAT scores.
Can a two point difference in LSAT scores really make such a big difference?
Yes, especially when there is a two point difference at the 50th and 25th percentile. Two points on a 120-180 scale may not seem like much, but the difference between 149 and 147 and 147 and 145 is actually very significant. In 2013, a 145 LSAT was at 26.8%, a 147 was at 33.7% and a 149 was at 40.5% (give or take a couple of tenths depending on when the test was taken). That 7% percentage point difference in ability often means the difference between bar passage and bar failure.
Let us assume that, on average, students at all three schools performed on the bar exam commensurately with how they scored on the LSAT, plus or minus a few percentage points. The average pass rate for first-time takers in Florida was 68%. Thus, the lowest scoring 32% of Florida bar exam takers failed. With thousands of bar takers from a wide range of law schools, we can assume that those taking the Florida bar exam represented the full spectrum of LSAT takers who were admitted to law school. Since virtually no one from those scoring in the bottom 10% of LSAT takers is admitted to law school, and only a small number from the next 5-10% (although this has been increasing in recent years), a student with a 147 is likely very close to the 28th percentile of LSAT-scorers taking the bar exam. Therefore, we would expect 147s to pass at about a 50% rate in Florida. Those below 147 would likely fail, unless they outperformed their LSAT score. At 145 and below, we would expect a very high failure rate. Starting at 148, a majority likely passed, with a higher percentage passing as LSAT scores increased.
Although Barry and St. Thomas had a 147 median, both schools experienced significant transfer attrition, with many of their top students transferring to more prestigious schools. Therefore the median of those taking the bar was probably slightly below 147. Therefore, it makes perfect sense that these schools had a less than 50% pass rate, whereas a school with a 149 median LSAT was over 60%. As another consistent data point Florida A&M University, which had a 147 median LSAT for their full-time entering class of 2013, and 148 for their part-time entering class of 2012, had a 2016 combined Florida first-time pass rate of 53.8%, with 49 passers out of 91 first-time takers; again, this is exactly what would be expected.
The upshot of all this is that schools in states with bar exams of comparable difficulty to Florida, a good rule of thumb would be that in order to maintain an acceptable bar pass rate, schools should not allow their median LSAT to drop below 149, or their 25% LSAT to drop below 147. (In states with tougher bars, like California, I would recommend adding a point. Schools in states with easier bars, like Missouri or Alabama could probably get away with going one point lower.) And schools should do everything possible to keep transfer attrition low and hold on to their highest-performing students, who are almost guaranteed bar passers. This still leaves room for law schools to admit up to 24% of very high (146-145 LSAT) and extremely high (144 and below) risk students in order to provide opportunities for deserving applicants who have some indicators of potential success. However, in order to guard against exploitation, law schools should separately track, and the ABA should require them to publish, attrition and bar passage statistics for the bottom quartile of each entering class by LSAT score. Reporting such results would also allow meaningful comparisons among law schools in their ability to prepare high risk students for the bar.