A law school's job placement success is one of the most important factors in a prospective law student's decision where (or whether) to attend law school, and it should be. Each year when the ABA Employment Outcomes disclosure is published, bloggers scrutinize the numbers to rank the performance of law schools in placing their graduates. And of course, the influential US News rankings incorporate placement success as a significant portion of the overall ranking.
Although many of these analysis are useful, most analysts commit the same error that US News does in analyzing this data. US News considers so-called "JD Advantage" jobs--those for which the JD is an advantage but not a requirement--the same as "Bar Passage Required" jobs--those for which bar passage is required (plus judicial clerkships). Equating these two categories of job placement outcomes is misguided and makes analysis of the data misleading.
There are two main reasons why the JD Advantage category creates problems. First, JD Advantage jobs on average are not "good" placements (although there are exceptions). Second, what qualifies as a JD Advantage job is vague, making the category highly manipulable. The combination of these two factors with the "full weight" of JD Advantage jobs in US News has made the category a serious problem for the integrity of placement success statistics. Because I don't think the susceptibility of the JD Advantage category to manipulation is terribly controversial, I will focus on Point #1 (that JD Advantage jobs generally are not good placements) to make my case in this post.
First, let's look at the category of placement that is an unambiguously "good" placement--Full Time Long Term Bar Passage Required Jobs. As is clearly seen in the plot below, schools with higher percentages of Bar Passage Required jobs have lower percentages of unemployment. That's what you would expect for two reasons: (1) schools that place graduates in "good" jobs versus "bad" jobs probably place more graduates overall, and (2) people with "good" jobs have jobs and therefore can't be unemployed.
However, let's look at the same plot for JD Advantage jobs. Here, the association between unemployment and JD Advantage is positive (rather than negative as in the plot above). That means that schools that have a higher percentage of unemployed graduates also have a higher percentage of JD Advantage graduates. That's surprising, because by definition unemployed graduates can't have JD advantage jobs, so we would expect a negative relationship even if JD Advantage jobs weren't particularly bad.
The fact that the relationship is positive is concerning for a system that equates JD Advantage jobs to Bar Passage Required jobs. Indeed, the plot above probably understates the how problematic JD Advantage jobs are for the reason mentioned above. To cancel out that effect, let's look at the number of JD Advantage jobs as a percentage of employed graduates (i.e., excluding the unemployed on the vertical axis), again against the percentage of unemployed graduates (on the horizontal axis). Here we see clearly that as the percentage of unemployed graduates for a law school rises, those who are employed are employed in JD Advantage categories at a higher rate.
The data shows strongly that JD Advantage jobs are not associated with "good" placement outcomes in general. That is not to say that all JD Advantage jobs are bad. There are a handful of JD Advantage jobs (probably mostly at elite institutions) in investment banks, consulting, and the like where the graduates could have easily secured a Bar Passage Required job. In most cases, however, large percentages of JD Advantage jobs are indicators of weak placement by the law school and possibly worse, given the inherent manipulability of the JD Advantage category.
I hope this post is enough to put those who analyze law school employment data on notice that JD Advantage jobs require more scrutiny than they currently receive. JD Advantage jobs clearly should not receive "full weight" in the US News formula, as this invites manipulation of job categories by schools and reduces the usefulness of the ranking. Sophisticated consumers of the data should see this category for what it is-- a highly manipulable disguise for generally low-quality placement outcomes.
For those interested in a more detailed analysis of these same data that suggests more appropriate weights for each job category, my previous analysis (and ranking of schools based on last year's data) can be found here.
Comments