People search for lists of schools with a 0% proficiency rate, and the number sounds damning: a school where not one student can perform at grade level. We pulled our data to see which schools those actually are, and the answer is more interesting, and much less scandalous, than the figure suggests.
Across the country, 52 public schools show a proficiency rate of zero in our analysis, with another seven landing above zero but below 1%. Out of roughly 100,000 public schools, that is a rounding error. And the schools in that group have a lot in common.
They are almost all very small
The single clearest pattern is size. Among the schools showing 0%, the median elementary school enrolls about 105 students, the median high school about 66, and the median middle school about 42. Some enroll fewer.
That matters more than anything else on this list. When only a dozen students take a test, a percentage stops being a stable measure. One or two students moving above or below the cut score swings the rate by 10 or 20 points. A single small cohort having a bad year can produce a zero at a school that posted respectable numbers the year before and will again the year after. The same math produces improbable 100% rates at tiny schools, and nobody calls those miracles.
States know this, which is why most suppress or withhold results for very small groups to protect student privacy. Those suppression rules create gaps in public data, and gaps sometimes surface downstream as zeros.
Several teach students who just arrived in the country
Look at the names in the group and a second pattern appears. Schools like Wellstone International High in Minnesota, Newcomers High School in New Jersey, and the Mission Education Center in California exist specifically to serve newly arrived immigrant students, many of whom are in their first months of learning English.
A student who arrived from another country in October and takes a state English exam in the spring is being measured on a language they are still acquiring. A low or zero proficiency rate in that setting describes the students' starting point, not the quality of the instruction.
The Oklahoma cluster is a reporting story
Most of the 0% schools in our data are small Oklahoma elementary schools, which almost certainly reflects how that state reports results for very small schools rather than anything distinctive about Oklahoma teaching. When a single state dominates a national extreme like this, the usual explanation is a difference in reporting or suppression rules, not a difference in classrooms. It is a good reminder that cross-state comparisons of raw proficiency are treacherous, since every state uses its own test and sets its own cut scores.
How to read a proficiency number
Three habits will keep you from misreading any school's score.
First, check the enrollment before you react to a percentage. If the tested group is small, treat the number as noisy and look at several years instead of one.
Second, ask who the school serves. A newcomer school, an alternative school, or a program for students with significant disabilities is not comparable to a comprehensive neighborhood school, and its raw scores were never meant to be read that way.
Third, remember what proficiency is. It is the share of students clearing one state's cut score on one test in one subject and year. It is a real signal, but a narrow one, and it correlates heavily with household income. Our explainers on how school ratings actually work and how to read your child's state test scores go deeper on what these numbers capture and miss, and school districts that beat the odds looks at what happens when you account for who a school actually serves.
Sources
Proficiency and enrollment figures from allk12's analysis of state assessment results and National Center for Education Statistics (NCES) data, 2024-25. Proficiency definitions, cut scores, and small-group suppression rules are set by each state and are not comparable across state lines.
