Imagine paying top dollar, sacrificing your personal life and compounding years of stress into mere months, all in pursuit of a goal you later decide to give up.
That's the situation many women face when applying to law school. After spending incredible amounts of time, money and effort to prepare for the Law School Admission Test (LSAT), they receive scores they feel are too low to get into the schools they want.
LSAT scores are often the most important aspect of a law school application, but people seem oblivious to the fact that men receive significantly higher LSAT scores than women.
Numerous news articles have bemoaned the decline of the number of women enrolling in law school over the past five years, blaming the dwindling number of female attendees on a slew of factors. Most blame a corporate legal culture that restricts family and personal time.
Yet these articles discount one critical fact: the percentage of women taking the LSAT during this same period has remained roughly static, and is higher than the percentage of women in law school. In 2005-2006, 49.08 percent of people taking the LSAT were female, while this year, law school is only 46.9 percent female.
The issue is clearly not that women are disinterested in law school, but that some post-exam event leads women to reconsider their original goals.
For many women, that event might be the receipt of their LSAT score.
Since my father is a law professor at the University of San Francisco, I asked him for some LSAT test data that he received from the school's admissions office.
Upon analyzing the data from the Law School Admissions Council for 2001-2006, I found the percentage of women who score in the top percentile of test takers - the select students scoring 175 or above - is roughly a third of a percent.
Around three-fourths of a percent of males scored in the same category. Using this measure, women finally outpace men in scores below 150 - a score lower than at least 75 percent of those admitted to each Tier 1 and 2 school.
If there were no gender differences, we would expect to see the same percentage of men and women landing in each bracket.
As much as it offended my feminist sensibilities, I stopped to consider whether men are better applicants than women and deserve higher scores. The LSAT is supposed to predict first-year grades in law school, so if the LSAT is an accurate predictor, we would see men obtaining higher GPAs than women in law school.
While I couldn't obtain law school GPAs, I could find undergraduate ones. When comparing test-takers' LSAT scores and undergraduate GPAs, I found that women had higher GPAs for each of the past five years.
Interestingly, while male and female GPAs both increased over the past five years, LSAT scores did not. The mean male LSAT score increased from 152.7 to 154, while the mean female LSAT score only increased from 151.1 to 151.5, making the differences greater in recent years.
I don't believe the correlation between greater gender differences and decreasing female enrollment is pure coincidence. I recently took the LSAT, and if my score hadn't met my expectations, I wouldn't have applied to law school. And I don't think I'm unusual.
With the media ignoring gender differences in the LSAT, women assume their individual performance - and not the system - is the problem.
The problem does not end with women deciding not to apply to law school. Possibly due to lower LSAT scores, they are also being admitted in lower percentages.
In 2006, which showed a difference in mean LSAT scores of 2.5 points, the number of women admitted decreased by 1.7 percent, while the number of men increased by 1.1 percent. This year had the greatest disparity of the years I analyzed.
Gender differences alone may not account for the decline of women enrolling in law school, but it is a factor that needs studying.
Based on my limited knowledge of neurological differences between men and women, my guess is that the LSAT emphasizes skills where men tend to have a natural advantage, such as certain spatial tasks, and deemphasizes skills where women tend to have stronger ability, such as short-term memory.
However, not being any sort of an expert in this field, I leave these differences to the actual experts to investigate and correct.
Otherwise, the number of women in law school will likely continue to decline, thus harming the diversity that is believed by the academy to be necessary for a "fair and balanced" legal education.
Colleen Honigsberg is an economic consultant in Washington, D.C. and former editorial page editor of The Daily Bruin.
Comments
comment
This is definitely well-writen and contains some very interesting analysis, but I have a couple of comments: "the percentage of women who score in the top percentile of test takers - the select students scoring 175 or above - is roughly a third of a percent." That's not quite true. By the time you hit 175, you're not looking at the top percentile. According to my September '07 score report, that's only half a percentile. Is the disparity between men and women at that level really statistically significant? How many people does that actually cover? (According to my quick calculations, around 3,000). It is over a five-year period, so I'm sure there's some merit to that analysis, but it did make me wonder. "Based on my limited knowledge of neurological differences between men and women, my guess is that the LSAT emphasizes skills where men tend to have a natural advantage, such as certain spatial tasks, and deemphasizes skills where women tend to have stronger ability, such as short-term memory." I think that's true. But are you suggesting that the LSAT be changed to incorporate skills more suited to women just to make it more fair to women? I don't think that the designers intended to bias the test towards men; they designed it to test skills needed for the law profession. Short-term memory, for example, is certainly less important to being a lawyer than the kind of reasoning seen in the LSAT. If the LSAT is indeed an accurate assessment of how good a lawyer a person will make (which is of course disputable), that leads to the politically incorrect conclusion that women don't do as well on the LSAT because their skills just don't lend themselves to being as good lawyers. I hate to say it, but your analysis might just mean that on the whole, men are naturally better lawyers than women. Of course, that's a pretty superficial conclusion on a complicated issue. In any case, very interesting column.
comment
Why is someone with no connection to Penn writing in the DP? Anyway, I could've summed this article up in one sentence: "Males tend to perform better on the LSAT, but I'm not an expert on this matter so I don't know why."
comment
Imagine the outrage if a MALE had written this: "...my guess is that the LSAT emphasizes skills where men tend to have a natural advantage, such as certain spatial tasks..."
comment
This is a fairly well written and researched article, but I have two comments. First, I can tell you from personal experience and the experience of my colleagues in law school that the LSAT does not necessarily reflect first year performance. I had friends who scored very low for my incoming class who made law review and friends who scored very high that barely graduated. I also scored lower on the LSAT than my wife, who I went to law school with, but graduated with a significantly higher GPA. So I for one think that theory is garbage. Second, you fail to differentiate among majors when considering GPA's. I'm not going to go into full analysis here, but I believe that in the engineering disciplines the ratio of men to women is far greater than in the humanities. And a 3.0 in engineering is not the same as a 3.0 in English. You probably should not have included the paragraph on GPA's if you were not able to obtain law school GPA data since the data you do provide is misleading. With that said, I think it is interesting that women are underperforming men on the LSAT and the growing gap should be studied further. Also you may not have data on the most recent LSAT exams, but I'm curious if the new format was meant to close that gap.
comment
Hello, to those of you who posted comments, thanks. As best as I can, I want to briefly clarify any confusion. ¥ ÒSecond, you fail to differentiate among majors when considering GPA's.Ó This is a really good point, and something I should have addressed in the article. I donÕt have GPAs associated with particular majors, but I do have the percentage of students coming from each major. There are definitely more men than women in both computer science and engineering majors. However, the number of students coming from these majors is low overall (in 2005-2006, CS Men: 1.96%, CS Women: .60%, Engineering Men: 4.51%, Engineering Women: 1.30%). Given that women consistently have average GPAs of roughly .10 higher than men, I think it is unlikely these students account for the entire difference, but I could be wrong. ¥ ÒAlso you may not have data on the most recent LSAT exams, but I'm curious if the new format was meant to close that gap.Ó The data that I have is from 2001-2002 through 2005-2006. I did see a study from the late nineties indicating that women were underperforming on the LSAT in comparison to their first-year law school GPAs, so itÕs possible the new format was meant to address this issue. However, if it was meant to close the gap, I think it has done a pretty poor job since the 2005-2006 year showed the greatest mean gender difference of any year I looked at. ¥ ÒBy the time you hit 175, you're not looking at the top percentile. According to my September '07 score report, that's only half a percentile.Ó Thanks. The word I meant here was Òbracket,Ó not percentile. I wrote ÒbracketÓ in my original piece, and then inadvertently changed it. ¥ ÒIs the disparity between men and women at that level really statistically significant? How many people does that actually cover?Ó For the 5 years of data that I have, there were 2,525 students who scored from 175-180. The gap between men and women is most pronounced in the 175-180 bracket, but is consistent throughout all scores from 120-180. Men are overrepresented in all brackets above 150, women in all brackets below 150. (A score of 150 corresponds to roughly the 50th percentile.) In determining significance, I was looking at the whole dataset. ¥ ÒBut are you suggesting that the LSAT be changed to incorporate skills more suited to women just to make it more fair to women?Ó I canÕt say at this point, but my instinct is that only a minimal change would be necessary for men and women to obtain statistically equivalent scores. My guess is that men and women generally perform the same on most questions, but that there are a few questions where men perform significantly better. For example, I would expect men to do better on the Òdinner tableÓ logical reasoning type of question simply because it involves mentally rotating the dinner table and the people sitting at it. There are ways to test this form of logical reasoning without requiring mental rotation of objects (a task where men outperform women and which I donÕt think is particularly important to being a lawyer). I think it would be fair to replace the types of questions that unfairly bias women (or men), but that donÕt test necessary legal skills.