You’ve just gotten your LSAT score back from the LSAC website after frantically hit F5 countless times on score-release day. For better or for worse, you now have a three-digit number that law school admission committees are going to give great consideration. On the other hand, maybe you don’t have your LSAT score and/or GPA yet, but want to have some idea of what you need to aim for to get into certain schools. Coupled with your GPA (as calculated by the LSAC/LSDAS), your LSAT score is going to play a significant role in determining where you will be admitted for law school.
Many applicants turn to admission prediction calculators to try to ascertain their chances. While Top-Law-Schools.com has a “What are my chances?” forum for applicants seeking a human touch and consideration of soft factors, there are a number of free online prediction calculators that should also prove useful to law school applicants during the admission process. It is important to keep in mind that no prediction calculator is perfect, and each tends to have its own quirks and weaknesses.
This article will examine five admission prediction calculators: Chiashu, HourUMD Law School Probability Calculator, Law School Probability Calculator (standalone), Law School Admission Council’s Search for Schools Based on UGPA and LSAT Score (also known as the LSAC Calculator), and Law School Predictor (LSP). While, strictly speaking, Law School Numbers (LSN) and Top-Law-Schools Stats (TLS) are not prediction calculators (and are not discussed at length in this article), many law school applicants use the information they contain to make their own predictions.
For the habitual cynics out there, it’s worth noting that the author of this article created the Law School Predictor calculator.Chiashu
Chiashu uses logistic regression based on entire applicant pool data (versus user-submitted data) in order to make its predictions. While its previous percentile-based predictions for users do not appear to be accessible at this time, Chiashu does feature school-specific graphs with a green “admit” line and a red “reject” line.
One of the more unique features of Chiashu is the Decision Dates graph available for each school; this feature helps delineate what month applicants were accepted or rejected at a given law school. The Decision Dates feature uses user-submitted data to Chiashu.
Chiashu is not particularly noted for being user-friendly. Registration is not mandatory, but being registered may be useful for applicants who wish to keep track of their applications on a site other than LSN. Without the percentile-based predictions that it used to have, Chiashu’s value as a predictive tool for law school applicants is limited largely to helping applicants get a general sense as to where they stand at a school, and the time of admission decisions at a given school.
Chiashu has not been updated recently, and no accuracy statistics in any form have been published for HourUMD.HourUMD Law School Probability Calculator
Published in affiliation with OurUMD.com, a website with web-based academic resources for University of Maryland students, the HourUMD Law School Probability Calculator relies solely on applicant self-reported LSN data in generating its results. By aggregating LSN data, HourUMD calculates the percentage of LSN applicants with similar numbers to the user who were admitted to a specific school and lists the result in the “Record” field. Furthermore, HourUMD also lists the percentage of applicants who got “In With Worse” numbers (both LSAT and GPA). If the user wishes, as the percentage of LSN applicants who received merit scholarship money (and the average award) will also be displayed in the results.
To better tailor the percentages to specific users, options for HourUMD include factoring in waitlisted LSN applicants, URM applicants (or only using URM applicants to calculate the results), and a rather unique option that allows the user to enter a range of LSAT scores and or GPAs for which Hourumd will obtain the LSN admit data (and scholarship info if selected).
HourUMD is arguably one of the best law school admission prediction tools that currently exist, and includes nearly all (if not all) ABA law schools as well a couple non-ABA law schools. It gives particularly meaningful results for applicants who are applying to top law schools and who have an LSAT score and GPA that fall between a school’s 25th- and 75th- percentiles, since the bulk of LSN applicant data also falls within these parameters. On the flip side, URMs, splitters (both traditional and reverse) and applicants who are interested in Tier 2 (weaker), 3, and 4 schools may find that there’s precious little data with which they can forecast their chances. In addition, the “In With Worse” has little predictive value for many splitters, since it is should be fairly obviously that having an LSAT score and GPA that fall below the 25th-percentiles (versus splitters which are or
) don’t tend to have much of a shot at being admitted.
Due to being self-reported (and from a self-selected pool), LSN data can also be subject to fudging, and it has often been suggested that the average LSN applicant has better numbers than the average law school applicant; in turn, this affects HourUMD calculations. Hopefully, it all balances out at schools with more applicants on LSN, but at the smaller/weaker-ranked schools, this could present some problems. In addition, HourUMD uses LSN data that can sometimes stretch back more than five years into the past. At a number of law schools, this should not have too much of an effect, but the top law schools have increasingly accepted only higher-and-higher numbers from applicants in recent application cycles, and thus HourUMD may be slightly over-optimistic at such schools.
No accuracy statistics in any form have been published for HourUMD, although HourUMD’s calculations change (and presumably become better) as more/new applicants create and update LSN profiles.Law School Probability Calculator (standalone)
This calculator appears to be the predecessor to HourUMD’s calculator, although there does not seem to be any definitive information that exists to indicate if this is true. It is basically the same as HourUMD except with far fewer features. Unlike HourUMD, this calculator does use logistic regression to generate a 95% confidence interval for admit rates which the user can see.Law School Admission Council’s Search for Schools Based on UGPA and LSAT ScoreAlso known as the LSAC Calculator
Another popular prediction tool, the LSAC calculator uses applicant data from the previous cycle to generate its predictions. By utilizing logistic regression analysis, the LSAC calculator generates a range of probabilities (based on a 95% confidence interval) of an applicant’s likelihood of admission. In a departure from other calculators, results are illustrated visually as a predictive green horizontal bar contained within a 0%-100% purple horizontal bar for each participating law school.
While the LSAC Calculator does contain a much larger dataset than HourUMD (or anything that would rely on LSN), the LSAC Calculator is limited its predictive utility for law school applicants. The prediction ranges are often quite broad, and there’s no supporting information offered to determine precisely how the prediction range was determined. Furthermore, a number of schools (notably top schools) do not opt to participate in the LSAC Calculator, and thus predictive ranges are not available for those schools.
The accuracy of predictions rendered by the LSAC Calculator is not published, but presumably are not too far off, given the wide range of predictions and access to the previous year’s entire applicant pool data.Law School Predictor (LSP)
Published under exclusive license to Top-Law-Schools.com, Law School Predictor is the new kid on the block when it comes to law school prediction calculators with the original version being released on December 14, 2008 followed by the official website launch on June 18, 2009.
Unlike other calculators, the foundation for LSP predictions relies on 25% and 75% LSAT and GPA information of matriculated students along with admission index formulas published by law schools (schools uses their own formulas) each year that take an applicant’s LSAT score and GPA and produce an “index score” that schools will use to group or and/or rank applicants. Regression analysis perhaps modified by a review of pertinent LSN data is used to create formulas for the handful of schools that do not publish them. Matriculated student and ranking data is gleaned from the most recent U.S. News & World Report: Graduate Schools Edition.
Like HourUMD, LSP factors in URM status in making predictions, and also is the only prediction calculator that incorporates a “binding early decision” feature to better calculate an applicant’s chances of admission. LSP includes school information for all ABA accredited full-time and part-time law school programs (full and provisional accreditation) and renders predictions for all but two of those programs (due to insufficient data) as of the start of the 2009-10 application cycle. Predictions are categorical and textual in nature, although a recent feature also indicates the estimated percentage of the last class of students who had index scores at or below the user’s index score.
Perhaps the most unique aspect of LSP is its behind-the-scene attention to details like whether an applicant is a splitter (high LSAT score relative to low GPA) or has a markedly low GPA. If such conditions are met, LSP will apply a hidden boost or penalty to an applicant’s chances as it relates to a specific law school. LSP is also the only prediction calculator that features a convenient LSDAS GPA calculator on its site to help applicants determine their LSDAS GPA and/or what grades they need to pull off in order to achieve a certain GPA.
One of the greatest drawbacks of LSP is its relatively long load times compared to the other prediction calculators. For a few law schools such as Berkeley, the published admission index formula does not appear to have a particularly strong relationship to actual admission decisions, and thus, LSP predictions for those schools can also be noticeably divorced from reality. While special algorithms are in place to adjust the splitter predictions, such algorithms remain a work in progress. Users have also occasionally expressed frustration that LSP does not calculate a specific percentage chance of admission, instead relying to textual prediction categories.
Accuracy statistics comparing LSN-listed applications (including a URM application subset) to LSP predictions are available on the LSP website, as well as a FAQ section and an About page that go into further detail as to how LSP works.