Data guide: how to read every number on this site

Law school data is easy to publish and easy to mislead with. This page explains what each field actually measures, which year it describes, and — just as important — which comparisons we refuse to make.

Last updated: October 2026

The short version. Every school-level figure comes from the ABA Standard 509 Required Disclosures (2025 reporting cycle), supplemented by official LSAC score-conversion data. We do not mix in third-party estimates, rankings, or modeled numbers. Where the disclosure does not support a comparison, we do not make it — and below is a list of exactly where that happens.

Why this page exists

Most law school data on the internet has a specific failure mode: numbers that are individually accurate but collectively misleading. A tuition figure averaged across schools that charge two different ways. A bar passage rate that quietly mixes cohorts. A “scholarship rate” that describes the students who got aid and not the ones who applied.

Each of those is arithmetically defensible and practically useless. Rather than ask you to take our word for it, here is every definitional choice we made, including the ones that make our rankings shorter.

Which year does each field describe?

The 509 is not a single snapshot. Different fields describe different academic years, and mixing them is one of the easiest ways to produce a wrong answer. Our datasets:

A 509 field describing tuition is a budget for an entering first-year student. It is not a bill you have already been sent.

Tuition: two reporting systems, and why we never average across them

ABA schools report tuition in one of two ways, and the difference is not cosmetic:

It is tempting to multiply per-credit tuition by an assumed number of credits to produce a “comparable” annual figure. We do not do this, and we would ask you not to trust anyone who does. The assumption is a guess, and guesses propagate: they distort totals, and they reorder rankings without changing a single fact about any school.

Per-credit schools are therefore listed separately and excluded from every total-cost ranking. The consequence is that 164 schools are cost-rankable, not 196. A school missing from a cost ranking on this site is missing because its reporting basis cannot support the comparison — not because it was overlooked.

Living expenses: what “on campus” really means

The 509 reports living expense budgets for on-campus and off-campus students. Two things routinely trip people up:

These two groups together cover all 196 schools, so every school page shows a living expense — with an accurate label about which kind it is.

Mandatory fees and the cost of attendance

Of the 164 annual-reporting schools, 152 report mandatory fees alongside tuition. The remaining 12 report no separate mandatory fee figure.

Those 12 schools still appear in cost rankings, calculated as tuition plus living expenses — and every such row is marked as a partial total. We never treat a missing fee as a zero: that would understate their cost of attendance and quietly hand them an unfair ranking advantage. A known-incomplete figure, labelled as such, is more useful than a complete-looking wrong one.

Because tuition, fees, and living expenses are the disclosed components of cost of attendance, the totals we publish are the cost of the first year. Multiply by three only with the caveat that tuition can change across your attendance, and that a JD is conventionally a three-year program — the 509 does not report program length.

Financial aid: two numbers that look contradictory on purpose

194 schools are covered by the aid data, and the eight aid fields are populated for every school we publish. Two of them describe different things:

These do not compose into a single number, and we do not build one. The median grant is a median over aid recipients only — it says nothing about what an aid non-recipient gets, which is nothing. A median grant is not an offer, and this site never presents it as one.

We also publish the four-band aid distribution, which is the only data here that separates need-based from merit-driven aid: schools with high full-tuition-grant shares give away far more, while schools with high “half or less” shares aid students who need it most. Two schools can both aid 90% of students and mean completely different things by it.

Conditional scholarships: where we exclude data on purpose

Conditional scholarships carry retention requirements — typically a GPA or class-standing floor. 60 schools report offering them.

Retention is computed from students who entered and students who lost the award. That denominator is small at many schools: of the schools reporting the detail, 7 have fewer than 15 students in the cohort. A single student's outcome moves such a figure between 0% and 100%.

So we publish those 7 separately and keep them out of the retention ranking. This makes our ranking shorter on purpose. It prevents one person retaining from outranking a school where hundreds retained, which would be indefensible as information.

Bar passage: the two rates answer different questions

This is the single most misread number in law school data, so we are explicit:

The roughly 7-point gap between them is the students who passed on a second or later attempt. Neither rate is wrong; they measure different things. If you are choosing a school, the ultimate rate describes the school's eventual outcome for its whole entering class. If you want the single hardest number, use the first-time rate. We publish both, always labelled, and never average them.

We also publish a third column: the school's two-year rate alongside its first-time rate, so you can see which schools rely on re-takes. Ten schools report an ultimate rate below their first-time rate, which is a real consequence of these being different cohorts. We keep the values exactly as filed and explain the discrepancy on the page rather than quietly adjusting either number.

Admissions funnel: yield and class fill are not the same thing

The four funnel fields — applications, offers, enrollees, and first-year class size — are complete for all 196 schools, and they support two different questions:

They can point in opposite directions, and reporting only one of them is misleading. A school can post a healthy yield and struggle to fill its class — or admit broadly and fill easily. Both figures appear together on every funnel page.

A low yield usually reflects generous admitting rather than an empty class, so do not read a low yield rate as a quality signal.

What we will not do

Known data quirks we have not corrected

Official disclosures are filed by each school and occasionally contain internal inconsistencies. We publish them as filed rather than editing them, because a silent correction would make the data impossible to audit against the source. The known cases:

Each is flagged on the affected school’s page at the point it matters, not only here.

Start with the data

Found a number that looks wrong? That is worth telling us about — tell us which school and which field and we will check it against the source disclosure.