Every Washington public school's state report card, in one place: ranked, trended, and compared.
Median school's share of students meeting the state reading standard
Washington publishes a lot of data about its schools, but it's scattered. Comparing a handful of schools means opening each school's report card one at a time, switching between measures and years, and reconstructing trends yourself.
I'm a parent with a daughter approaching kindergarten age. I built Compare Oregon Schools to pull every Oregon school's report cards into one place, and this is the same tool for Washington, built from the Office of Superintendent of Public Instruction's public data files. Rank schools, see how they've changed over time, and compare any schools you like.
This is an independent project, not affiliated with or endorsed by OSPI. Test scores and attendance capture only part of what makes a school good, so use this to narrow your list, then go visit. Check a school's official report card before making a decision.
Test scores are the percent of tested students meeting the state standard (level 3 or 4 on the Smarter Balanced tests, which OSPI now calls "consistent grade‑level knowledge and above"), across all tested grades at a school: reading (ELA) and math in grades 3–8 and high school, and science in grades 5, 8 and 11. High school reading and math were given in grade 11 through 2016–17 and in grade 10 since. A K‑5 school's score reflects grades 3–5; a K‑8 school's includes middle school. Students who take the alternate assessment for students with significant cognitive disabilities (WA‑AIM, about 1% of students) are reported separately by OSPI and aren't included. Reading & math combined is the simple average of the two. Science is left out of it because it's tested in only one grade per school level and changed tests in 2017–18.
The Washington comparison line for each school is the statewide rate in the grades that school tests, weighted by how many of its students tested in each grade.
"86% tested" means fewer than 90% of eligible students took the test in at least one year shown; the tag gives the lowest rate and, for multi‑year views, the year it happened. Students who opt out may differ from those who test, so the score may not represent the whole school. Where fewer than 40% tested, the score isn't shown at all. This matters most for high schools in 2014–15 to 2016–17, when many 11th graders skipped the test. Participation is tracked for all students only, not student groups.
"±13" is a 95% margin from sample size alone. A school with 40 tested students carries roughly ±15 points of noise in a single year, so small differences between small schools mean little. The 3‑year average helps.
"few tested" appears in adjusted mode when a school averages fewer than 20 tested students a year; its adjusted score has been pulled toward zero (see below). "charter" marks Washington's charter public schools, each of which is its own district; they enroll by application and lottery, which tends to select for engaged families, so their results aren't directly comparable to neighborhood schools. Use the checkbox to include or exclude them from the list and charts. "online" marks online and virtual schools, identified by name (for example "Online Academy" or "Virtual Academy"); they have their own checkbox. The list covers regular, alternative, tribal compact and contract schools; juvenile detention centers, state institutions, re‑engagement programs, skills centers and special‑education centers are left out.
Why the charts plot schools against poverty: not to point anyone toward low‑poverty schools, but to separate what a school adds from what its students bring with them. Raw test scores mostly track family income, since children from families with more resources arrive at school with more support. Comparing a school with others serving similar students shows which ones do more than expected with the children they have, which is closer to what a parent is choosing when they choose a school. A school well above the curve at 50% poverty may be doing more for its students than one with higher raw scores in a wealthier neighborhood.
Poverty here is OSPI's low‑income rate: the share of students from families eligible for free or reduced‑price meals, from each year's October enrollment count. This fits a curve of each measure against that rate, across every Washington school of the same type that year, and ranks schools by how far above or below that curve they sit. The curve is a quadratic constrained never to rise with poverty; a straight line fit noticeably worse and flattered the lowest‑poverty schools. Small schools are handled with a standard statistical adjustment (empirical Bayes shrinkage): each school's distance from the curve is pulled toward zero in proportion to how noisy it is, based on how many students tested. A school with 300 tested students keeps nearly all of its result; one with 15 keeps a fraction. The adjustment controls for poverty only; selective schools tend to look better than their true effect.
Student groups show results for one group of students, such as low‑income students, at every school. Comparing the same group across schools is one of the fairer ways to compare schools. Reading, math, attendance and on‑track results for groups cover the latest three school years; graduation covers the last three graduating classes. OSPI hides results for groups of fewer than 10 students.
District figures combine each district's schools, so they follow the grade and charter filters: choose Kindergarten to compare districts' elementary schools only. Test scores are weighted by students tested, most other measures by enrollment, and enrollment counts are summed. Graduation and on‑track rates use OSPI's official district figures instead, since district cohorts include students in programs without their own school results; those don't change with the filters. Schools whose results OSPI hides because they're very small drop out of the totals.
Regular attendance is the share of students who missed fewer than two days a month on average, for any reason. 9th graders on track is the share of 9th graders who passed every course they took. Graduation is the four‑year cohort graduation rate and five‑year graduation the extended rate that also counts students who took an extra year, both shown in the year students finished. Suspensions and expulsions is OSPI's discipline rate: the share of students suspended or expelled at least once during the year. Support staff and AP courses come from the federal Civil Rights Data Collection, gathered every other year (2017–18, 2021–22 and 2023–24 shown); support staff counts counselors, nurses, psychologists and social workers per 500 students. In 2023–24 most schools without AP courses left the AP question blank, so that year shows only schools that offer AP. Some high schools emphasize International Baccalaureate, Running Start or College in the High School instead of AP, which the AP count doesn't capture.
Enrollment change and kindergarten enrollment come from OSPI's October 1 enrollment counts; change is measured from fall 2019, the last fall before the pandemic. A shrinking school may face program cuts, consolidation or boundary changes.
This is OSPI's per‑pupil expenditure for each school: what the school spends directly plus its share of district‑wide costs such as administration, facilities, transportation and benefits, divided by enrollment. It varies for reasons that have little to do with quality. Money follows need by design, so schools with more low‑income students, students with disabilities or English learners receive more; small schools cost more per student because fixed costs are spread over fewer children; and some schools house district programs, such as special‑education centers, that raise their figure.
Washington's charter schools are each their own district, so their figures include their own central costs and are comparable with district schools; they're included. Online schools report far lower spending than buildings with classrooms, so they're marked "not comparable" in the ranking and left out of the spending chart. A few schools report figures more than three times, or less than a third of, what schools with similar students spend; these usually reflect how a district's books are split rather than what happens in the building. They're marked "unusual figure" and left out of the chart.
The spending chart compares each school with schools serving similar students. Across: how much more or less it spends than a school with the same poverty, special‑education and English‑learner shares and size typically spends statewide. Up: its reading and math results relative to what its poverty level predicts. Dividing results by dollars would mostly reward cheap, affluent schools, since money flows toward need; comparing like with like avoids that.
There were no regular spring state tests in 2019–20 or 2020–21 (a shortened test given in fall 2021 isn't comparable and isn't shown). Science changed from the MSP to the WCAS in 2017–18, and high school science starts that year. Rates OSPI hides behind a range such as ">90%" or "<5%" are plotted at the middle of that range.
Sources: Washington Office of Superintendent of Public Instruction report card data on data.wa.gov (assessment, enrollment, attendance and on‑track, graduation, discipline and per‑pupil expenditure files); U.S. Department of Education Civil Rights Data Collection. All public data.