December 2025 / Episode 6 / Under 15 minutes
Measuring the LSAT's Predictive Validity
Welcome to Keeping Up to DataSM, a space in which we discuss, analyze, and contextualize trends and perspectives in the current law school admission cycle.
SUSAN KRINSKY: Welcome back to Keeping Up to Data. I’m Susan Krinsky, executive vice president for operations at LSAC®, with an update on the 2026 application cycle and an interesting discussion with our chief assessment scientist, Anna Topczewski, about the predictive validity of the LSAT® and the correlation studies that provide the data that enables us to measure the LSAT’s predictive validity.
It’s just over a week into December as I look at our volume data. Last year at this time, we had seen 42% of the applicants who eventually applied and we had received 40% of the applications — so, not quite halfway through the cycle, on the basis of applicant and application volume. As of today, the number of individuals applying to law school is 21.7% higher than last year at this time and 52.7% higher than two years ago. Looking at applications submitted so far, it’s 62.7% higher than two years ago, but 22% higher than last year at this time.
The increase over last year continues to moderate a bit, as it often does as we move through the annual cycle. As of this week, 46% of individuals who have submitted at least one application for 2026 admission identify as persons of colo;. 57% identify as female; 40% as male; 1.8% did not indicate a gender; and just under 1% identify as gender diverse. 24.6% of applicants described themselves as being first-generation college attendees or graduates. Applicants from all regions of the U.S. are up, with applicants from the Mountain West and Northwest up the most, over 30%. Canadian applicants and applications are up 22.6% and 23.8%, respectively.
We’re now halfway through the testing cycle, having completed four of the eight LSAT administrations. Test takers for the first four LSAT administrations of this cycle — August, September, October, and November — were up 19% over the first four test administrations in 2024, with over 112,000 test takers so far. First-time test takers over those first four test administrations are up 11%. But it’s probably worth noting that the November 2025 test had the highest percentage of repeaters, 64%, that we’ve seen over the past three years. The next LSAT administration will take place in January, and as of 30 days before the administration begins, registrants are up 8% over last year.
That’s our report on the December data. As you know, you can stay on top of these numbers by going to our website, LSAC.org, where you’ll find a link on the homepage to the latest volume data, which is updated every night.
Joining me today is LSAC’s senior director of assessment sciences, Anna Topczewski, and we’re going to talk about the correlation studies that LSAC produces each year. Anna and her team are responsible, among other things, for completing the analyses that go into the correlation studies, which compare LSAT scores and undergraduate GPAs to first-year performance in law school. Participating schools are provided with school-specific results — that is, how well do their students’ LSAT scores and undergraduate grade point averages predict performance in the first year at that school? Asked another way, how strong is the predictive validity of the LSAT, the undergraduate GPA, and the LSAT and undergraduate GPA together?
Anna’s team also produces a periodic summary report combining several years of correlation studies, the most recent of which summarized five years of correlation study results, from 2020 through 2024. Anna is with us today to explain the concept of predictive validity and to talk to us about the report’s findings. Anna, welcome to the Keeping Up to Data podcast.
ANNA TOPCZEWSKI: Happy to be here.
SUSAN: Before we get into the concept of predictive validity, Anna, why don’t you tell our audience a little bit about the background of the correlation studies? When did they start, and why are they done?
ANNA: For more than seven decades, LSAC has worked to provide correlation studies to examine how well the LSAT predicts success in law school. Any admissions test has the purpose of differentiating who is likely to perform well from those who are less likely to perform well, and the test needs to fit the purpose. Does this test actually predict what it is supposed to predict?
In the end, predictive validity is a fancy term for how well a test correlates with the outcome it is supposed to predict. The purpose of the LSAT is to get a sense of how well prospective students will perform academically in law school. If the LSAT scores are highly correlated with law school grades, the test demonstrates strong predictive validity. And, spoiler alert, the LSAT has been and continues to be a strong predictor of success in law school.
SUSAN: So, where does one even start in the process of completing predictive validity analyses?
ANNA: It starts with individuals taking the test. When the law school admissions process happens, individuals are admitted to and enroll in a particular law school, and those individuals complete their first year. After that, law schools send us their students’ first-year grades; then, we begin the analyses and conclude by sending participating law schools their own, unique report. Participating law schools send us their students’ first-year grades every year, and we complete the analyses and reports every year.
SUSAN: Since we know that law schools use a holistic admissions process taking many factors into account, why are other things, like strength of letters of recommendation, personal statements, work experience, and the other information in the application file, not included in the predictive validity analyses?
ANNA: On the surface, the simple answer is, the LSAT scores and the undergraduate GPA are the “number-y” things in an application. They are the quantitative factors in the same way the first-year law school GPA is quantitative. LSAT scores are on a range of 120 to 180. In the application process, undergraduate GPA is standardized to the same GPA metric. LSAT and undergraduate GPA are the quantitative values that can be correlated with first-year law school GPA.
Letters of recommendation and personal statements provide context: who an applicant is. What are their life experiences? These are not innately quantitative things, and in many cases, nor should they be. And so, if it’s not a number, we can’t include it in the predictive validity study.
SUSAN: For context, the majority of law schools participate in the correlation studies by sending LSAC their students’ de-identified first-year grades, and LSAC not only examines the predictive validity of the LSAT scores, but also undergraduate GPA. Could you talk us through that a bit more?
ANNA: The LSAT correlation studies look at how well students’ first-year average in law school is predicted in several ways. First, we look at the relationship between LSAT score and first-year average. Second, we look at the relationship of undergraduate GPA and first-year average. And finally, we look at the combination of LSAT score and undergraduate GPA and how well those two factors together relate to first-year average.
But how we look at LSAT scores also varies in two ways, because individuals can take the LSAT several times. So, we look at the average of their scores, and we also look at the highest score an individual has achieved. We do this because we know different law schools look at LSAT scores in these ways, so our analyses need to align with what the law schools are doing in their admission process.
SUSAN: There have been some questions about how the replacement of the analytical reasoning section of the LSAT with a second logical reasoning section impacted the LSAT’s predictive validity. What can you tell us about what you know so far?
ANNA: The timeline of steps is important in answering this question. This recent correlation report provides summaries of the results for law schools participating in the 2020 to 2024 LSAT correlation studies. Therefore, these correlation studies represent a large span of LSAT test takers from about 2015 to 2023. Why the lag? Back to what I said earlier: These correlation analyses need law school students’ first-year average. So, let’s walk through the timeline of when we will start to get these data on those individuals who took the LSAT without the analytical reasoning section.
The test changed starting in August 2024, at the beginning of the 2024-2025 testing year and application cycle. Those who took the LSAT without the analytical reasoning section could only first apply during the just-passed admission cycle — that is, fall of 2024. And if they were accepted, they are now in their first year of law school right now. So, we wait for them to complete their first year of law school, and after they do, we will complete the predictive validity analyses, as we have done for over 70 years.
I would expect the predictive validity of the LSAT will not change as these results come in. I can say that because the LSAT has not changed in what it measures. The LSAT measures reasoning and reading comprehension skills. Yes, the LSAT format has changed several times as we went through and came out of the COVID pandemic, and while the analytical reasoning section was removed, what remained constant was the skills being measured. The analytical reasoning and logical reasoning sections differed in how they measured reasoning skills, but what they measured was the same. Because the skills of reasoning and reading comprehension have remained constant in the LSAT, I would expect that the LSAT will continue to be a strong predictor of first-year success in law school.
SUSAN: The latest LSAT correlation study summary report, which is available on LSAC.org, was released in late May. What findings stood out to you?
ANNA: First, I think it’s important to say what these data are and what they are not. In the report, you will see trends of LSAT scores and undergraduate GPA. These are trends of those who have matriculated into law school; these are not trends for all LSAT test takers or applicants. And the trends across these three groups are not always the same, especially for LSAT test takers. We’re only talking about individuals who have actually enrolled in law school and who have completed their first year.
Now, on to findings. And in many ways, the results are a bit boring, and what we have found for years is what we found in this latest report. It’s worth noting that this report covers the predictive validity trends over several changes in the delivery of the LSAT and the content of the LSAT. The LSAT score is a stronger predictor of success in law school than undergraduate GPA, but the LSAT score and undergraduate GPA together provide the best prediction of law school first-year average. We have also found that for many law schools, but not all, using average LSAT, rather than the highest LSAT score, is a better predictor. And generally, the difference in the predictive power between the average or highest LSAT score is quite small. Law schools can refer to their own, unique correlation study results, to see what we have found based on their own data, and tailor their admission process as they see fit.
SUSAN: The admission cycle, which closed this past August, saw 18.3% more applicants than the year before, and based on the number of test takers at the end of last testing year and the start of this testing year, we are trending toward another strong admission cycle this year. Anna, how does the increase in the number of test takers affect the LSAT’s predictive validity?
ANNA: In many ways, the volume of LSAT test takers does not impact the LSAT’s predictive validity. The statistic to monitor is the variance of the LSAT scores. When there are shifts in the volume of test takers, it’s important to look to see if the test takers’ scores are clumping (getting less variable) or expanding (getting more variable). In predictive validity studies, when the predictor clumps, it usually leads to less predictive validity. But what we have found is that the LSAT scores have been trending to being slightly more variable with the increase in test taker volumes. So, we expect LSAT scores to continue to be a strong predictor of success in law school.
SUSAN: Thank you, Anna, for joining me today, and for explaining what the correlation studies are and how they tell us about the LSAT’s predictive validity.
ANNA: Thank you, Susan, for having me here today.
SUSAN: Thank you for joining us at Keeping Up to Data. We look forward to your joining our next episode. Until next time, stay well.
Thank you for joining us. Keeping Up to DataSM is a production of LSAC. If you want to learn more about the current law school admission cycle and the latest trends and news, visit us at LSAC.org.