Where our numbers come from

School names, addresses, enrollment counts, grade ranges, student-teacher ratios, and demographic breakdowns come from the NCES Common Core of Data (CCD), the U.S. Department of Education's comprehensive annual directory of all public elementary and secondary schools and school districts.

Test score data comes from two sources: state Departments of Education (higher quality, most recent year) and federal EdFacts data from NCES (multi-year historical data for growth calculations).

Neighborhood statistics (median income, home values, education levels, commute times, poverty rates) come from the U.S. Census Bureau's American Community Survey (ACS), matched to each school's ZIP code.

Athletics, advanced STEM courses, and gifted and talented indicators come from the Civil Rights Data Collection (CRDC), a biennial survey of all public schools conducted by the U.S. Department of Education's Office for Civil Rights. We access this data through the Urban Institute Education Data Portal. AP/IB rows can come from CRDC, NCES EdFacts, or both; school pages show the row's recorded source.

Programs data is shown on school pages as supplemental information. It is not factored into our composite scores or rankings. Private schools are not covered by CRDC, so athletics and advanced course data is only available for public schools.

Our 1-10 composite ratings are calculated by MySchoolScout using the data sources above. The formula blends student growth, poverty-adjusted academic performance, and learning environment — plus college readiness for high schools — with weights that vary by school level. Equity data is shown for context but is not part of the composite. Rankings are computed within each state to ensure fair comparison across different state assessments.

For full details on how ratings are calculated, see our Methodology page.

On some school pages, the short summary under the school's name is written by a language model running on MySchoolScout's own hardware. It writes only from that school's facts: the same directory, rating, test score, program, and graduation data the page shows. Every number, name, and program it mentions is checked against those facts before the summary can appear.

A summary that fails the check is dropped, and the page shows a sentence composed directly from the data instead. School pages without a written summary use that composed sentence too.

Some school pages include a "What's said publicly" section synthesizing what a school says about itself and what independent sources record. We pull from the school's own website, its Wayback Machine snapshot (so we capture the page as a parent would have seen it last year), and Wikipedia when an article exists. Every claim is grounded in a cited source — we never include claims that don't appear in the underlying text.

Claims that fail the grounding check are pruned before render. Schools without a confident Wikipedia match are still synthesized using the remaining sources.

Government education data has inherent publication lags. The NCES CCD directory is typically released 12-18 months after the school year ends. State test score data is usually available within 3-6 months. Census ACS 5-year estimates are released about 12 months after the survey period.

This means some data on the site may be 1-2 years old. School pages show the held period when it is known and say “Vintage not available” when it is not. Refreshes follow each source's acquisition schedule, and a known newer release is disclosed until it has been validated and loaded.

Some data is suppressed to protect student privacy. When fewer than 10 students are in a group, state agencies suppress the data. This means small schools may have incomplete test score records.

None of these states becomes zero or an unfavorable rating. School facility, Census tract, ZIP/ZCTA, district, city, and county are also distinct geographies. Area context never proves attendance eligibility; verify zoning with the responsible district.