October 15, 2008
LSATs, SATs, Sound and Fury
Posted by Christine Hurt

'Tis the season for many things, but mostly silly ranking things.  Like others in the blogosphere, I received my U.S. News & World Report survey as the most recently tenured member of my faculty.  So that we may positively influence others' "Peer Reputation" scores, we are all receiving "law porn," whether we are among the four survey participants at our school or not.  Admissions deans are gearing up for a new admission season so that they can positively influence the LSAT scores, GPA scores and Selectivity scores.  One begins to think that our core Fall mission revolves around the USNWR.  To that end, three stories have been in the news in recent weeks about admissions and alleged gaming of the rankings systems.

The LSAT Stories:  Michigan and Illinois

First, Michigan was in the news about its proposed 3 + 3 program whereby talented undergraduates at Michigan will be admitted to the law school after three years of undergraduate performance.  These applicants will be accepted early in their careers before they take the LSAT; indeed, applicants may not have taken the LSAT.  On the one hand, one might say Hallelujah!  For these high-achieving undergraduates (must have 3.8 to be considered), about whom we have a great deal of information, being familiar with their majors, coursework and grading systems, we do not need your silly LSAT signal.  Others, of course, say that this is Michigan's way of gaming the rankings, by eliminating some LSATs, which may be low (or high) from its pool of LSATs.  (For selected coverage of the controversy, see Paul Caron's roundup here, Bill Henderson's take here and Tom Bell's here.)  Last week, Illinois (my institution) announced its own "iLEAP" plan that will also consider its own undergraduates with at least a 3.0 early in their career, before they take (and so long as they do not take) the LSAT.  (UPDATE:  Our admissions officer corrects me on this point -- applicants are not barred from the taking the LSAT.)  Note that although some have described the program as admitting students with a 3.0 and above, that is not the case.  Students will be considered with a 3.0 and above.)  Not to discriminate, controversy ensued again.  (See here, here and here.)

I don't know if I've ever met anyone in legal academia who believed that strongly in the LSAT or, in particular, in the ability to judge the quality of a law school on the LSAT score of the student whose score was situated on the median of her classmates' LSAT scores.  I've always thought that most academics were enlightened enough to know that how one applicant scored on a standardized test one Saturday only gave us the roughest proxy for how well that person would do the first year of law school, how well that person would interview with future employers, how well that person would represent an actual client, and so on.  I assumed that we all agreed that LSATs became a focus once we were ranked on that basis; if USNWR ranked us all on the basis of students' batting averages or attendance in grade school, then we would focus on those statistics instead.  Knowing this, I naively thought that any program that might allow a student to be judged holistically, without this very rough signal, might be applauded.  Especially at a state institution with a special claim to educating people of that state, some perhaps the first professionals in their families.  Given the fact that LSAT is influenced by many factors of wealth and socioeconomic class, we might think that such a program would be a small step toward levelling the playing field.  I have no idea whether the small number of students admitted in either program will affect an institution's median LSAT.  I also don't know if we can predict whether these high-achieving students would have done so poorly on the LSAT that we wouldn't have wanted to report them.  But I like any small step away from LSAT-centric admissions.

The SAT Story:  Baylor University

Today's NYT reports that Baylor University offered admitted freshman last year a credit of $300 at the college bookstore if they would retake the SAT (for free).  If any freshman raised her score by 50 points or more, then that student would also receive a $1000 merit scholarship.  Of the 3000 freshman that matriculated at Baylor, 861 students retook the SAT and received the bookstore credit.  The article reports that 150 students qualified for the scholarship; whether a total of 861 students retook the SAT or 1011 retook the SAT is unclear.  The average SAT score of incoming freshmen was raised from 1200 to 1210 for USNWR purposes.  So, there are cries of "misuse" and "manipulation."  University officials say that merit scholarships are based on SAT scores, so the retesting gives more students more chances for scholarships.  (Of course, Baylor could change its criteria for handing out merit scholarships, but let's bracket that for this discussion.)

So, is the takeaway lesson from this story that Baylor (a Baptist university) evil?  No.  The takeaway lesson is that the SAT is flawed, or at least ranking schools based on the average SAT scores of their students is flawed.  If any given applicant can raise his SAT merely by taking it a second time (not by taking a prep class, etc., but merely retaking it), then SAT scores don't seem that reliable -- particularly if over 17% can increase their scores by 50 points or more.  The highest SAT scores may merely go to those that have the luxury of time and resources to take the SAT multiple times.  Baylor is not a better university this Fall merely because less than a third of its freshman retook a test.  Baylor is the same university.  If we randomly retested 100% of all the freshmen at a handful of colleges, would they all jump in the rankings?  If so, then the SAT is evil.

Law Schools/Lawyering | Bookmark

TrackBacks (0)

TrackBack URL for this entry:

Links to weblogs that reference LSATs, SATs, Sound and Fury:

Recent Comments
Popular Threads
Search The Glom
The Glom on Twitter
Archives by Topic
Archives by Date
January 2019
Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
Miscellaneous Links