The standardized testing landscape in K-12 has been quietly fragmenting for the past several years. The Common Core era of relative state-level alignment has given way to a more diverse mix of state-specific tests, retained pieces of older systems, and a handful of states experimenting with through-year and adaptive models that look fundamentally different from the spring-test model. Some of the changes are policy-driven; some are reactions to pandemic-era learning loss; some are political reactions to what testing represents culturally.

For administrators, parents, and education observers trying to follow the policy thread, here’s the honest state of state testing in 2026.

What’s been kept

The federal foundation has not gone away. The Every Student Succeeds Act, passed in 2015, requires states to test students in math and English Language Arts annually in grades 3 through 8 and once in high school, and in science at three points across grades 3 through 12. The federal requirement is binding; state-level changes operate within the boundaries it sets.

Most states are still administering tests within roughly that structure. The names vary (Smarter Balanced in some states, PARCC’s successors in others, MCAS in Massachusetts, STAAR in Texas, Regents in New York, and a long tail of state-specific tests), but the architecture is similar: spring-administered, paper-or-online tests, results coming back over the summer or early fall, used for school accountability ratings.

The accountability framework is largely intact. Schools and districts are still rated based on student performance and growth on these tests, with various state-specific weighting. The use of test scores to identify schools needing intervention continues, even where the specific labels and consequences have softened.

The state-mandated pre-K through 12 testing infrastructure has therefore not collapsed despite a decade of cultural pushback. The structure exists; the question is what’s being done within it.

What’s been dropped

Several specific things have receded.

SAT or ACT as a state-required graduation or assessment test. Several states that had moved to using a college-admissions test as the high school assessment (Michigan, Illinois, Colorado in various forms) have either dropped the requirement or softened it. The arguments for the change varied by state, but the practical result is that the SAT and ACT are less universal as state-mandated tests than they were five years ago. Whether this matters for individual students depends on whether their college plans involve test-optional or test-required schools.

High-stakes graduation tests in some states. Massachusetts dropped MCAS as a graduation requirement in 2024 (though the test continues for accountability purposes). New York paused the Regents graduation requirement during the pandemic and has been reworking it since. The trend across several states has been to keep the testing for measurement and drop the high-stakes consequences for individual students.

The “Common Core era” cross-state alignment. Smarter Balanced and PARCC, the two consortia that emerged from Common Core, have shrunk meaningfully. Several states have left both. The result is less direct comparability of test scores across states, which makes national-level comparisons harder and state-level conversations more isolated.

What’s changing

A few experiments worth watching.

Through-year testing. Several states are piloting models where the annual assessment is broken into three or four shorter assessments across the school year rather than one long spring test. The argument is that the data is more useful to teachers and that the testing burden on any one day is reduced. Early results are mixed; the implementation matters as much as the design.

Adaptive testing. Smarter Balanced and several state tests have used computer-adaptive formats for years. The newer experiments are pushing further: items that adjust within a single session based on student responses, item banks that are larger than any single student would see, and reporting that focuses on diagnostic information rather than score-level summaries.

Innovative item types. Tests that include essay-grading by AI (with various levels of human oversight), interactive science simulations, and real-world performance tasks have grown. The validity research is uneven; the adoption is happening anyway in some states.

Reduced testing time. Several states have shortened their spring tests, sometimes substantially. The trade-off is reduced reliability of individual student scores in exchange for less classroom time spent testing. The trade-off is real and reasonable people disagree on the right balance.

The chronic absenteeism complication

One specific issue worth flagging: the elevated chronic absenteeism rates are interacting with state testing in ways that have not been fully reckoned with. State accountability systems require participation rates above 95 percent. Districts in high-absenteeism areas struggle to hit this. The data that does come in is from a non-random subset of students. School ratings based on this data are less reliable than they were pre-pandemic, and the policy infrastructure has not caught up to that fact.

Some states are quietly adjusting accountability calculations to account for participation challenges; others are not. This is one of the more consequential under-discussed pieces of the current testing conversation.

What this means for districts

For administrators thinking about testing in 2026, three observations.

State-level requirements are still real. Federal compliance, state accountability, and parent expectations create a floor that districts cannot opt out of, even when they would prefer to test less.

The space for local choice has grown. Within state requirements, districts have meaningful latitude in how they prepare for tests, what supplementary assessment they do, and how they use the results. The “pacing the curriculum to the spring test” mode that defined a lot of the 2010s has loosened in many places.

Test prep as a major instructional activity has receded. Districts that built heavy test prep programs in the early 2010s have mostly trimmed them. The evidence on test prep effectiveness was always thin; the cultural pressure has eased; the time has flowed back to instruction.

What this means for parents

For families navigating the current landscape, a few practical points.

The test results matter more for school accountability than for individual student decisions. Parents who treat their child’s state test scores as the primary signal of academic standing are usually overweighting them; teacher observations, classwork, and broader assessment patterns are more reliable for an individual kid.

Opt-out provisions exist in some states. The legality and consequences vary; in states where opt-out is a legal option, families exercising it should understand both the principle and the practical consequences for the school’s accountability data.

For high-school-bound homeschoolers, the state testing question is different. Many homeschool families take tests outside the state system for college-application purposes (SAT, ACT, AP exams) regardless of what the state requires. The state-by-state homeschool reporting question overlaps with the testing question in some cases.

The bigger picture

The state testing infrastructure is doing roughly what it has done for the past two decades, with some softening of stakes, some experimentation in formats, and some acknowledgment that the pandemic complicated the data. It has not been transformed; it has not been abandoned; it has been moving slowly in directions that are easier to describe state by state than as a national story.

For the foreseeable future, state tests are part of the K-12 fabric. The conversation worth having is which uses of the data are valid (informing instruction, identifying systemic issues), which are weaker (high-stakes individual student decisions), and how the system can be improved within the constraints. The argument about whether to test at all has mostly resolved in favor of testing; the argument about what to do with the results is ongoing and worth continuing.