The New York State Education Department is administering new Regents exams in Chemistry and Physics for the first time this June, replacing the prior format with assessments fully aligned to the NY P-12 Science Learning Standards — standards built on the same National Research Council framework underlying the Next Generation Science Standards. Every question maps to one of three dimensions: Scientific and Engineering Practices, Disciplinary Core Ideas, and Crosscutting Concepts. Approximately 60 percent of each exam is multiple choice; 40 percent is constructed response. Students cannot pass by recalling facts alone. They have to apply a practice — modeling, data analysis, argumentation — to reason about a core scientific idea.
For science teachers outside New York, this is where national assessment is heading. New York's first administration is the largest real-world test of what three-dimensional science assessment looks like at scale — where it stresses traditional curriculum coverage, what it demands from students in writing, and what teachers need to have actually built into instruction rather than mentioned once in a unit. States that have adopted NGSS-aligned standards are watching this closely. The question is whether curriculum kept up with the standards shift, or whether students arrive at test day having been taught the core ideas without having practiced the reasoning.
On May 12, 2026, the Michigan Department of Education announced AI guidance for K-12 schools, releasing both a Starter Guide and a Comprehensive Framework. State Superintendent Dr. Glenn Maleyko framed it plainly: AI can be a useful learning and teaching tool when used properly, and the guidance will continue to evolve as the technology changes. The framework identifies six essential practices for districts: keep AI purposeful and safe, protect privacy and integrity, build AI literacy, maintain human oversight, promote equity and accessibility, and support transparency and continuous improvement. Michigan joins a growing list of states issuing formal frameworks. Nationally, approximately 80% of districts now have some form of AI policy, up from roughly 57% in 2025.
The remaining roughly 20% of districts without AI policy are increasingly isolated, and the pressure to close that gap will only build through 2026-27 as AI-related academic misconduct cases continue to reach school boards. For teachers already in districts with policy, the more relevant question is whether the policy is functional or decorative. Michigan's two-document approach — a short Starter Guide alongside a longer Comprehensive Framework — is a design worth noticing. Districts that hand teachers a 40-page framework without a clear entry point will get the same compliance theater they get with any other unfamiliar mandate. The ones that start with a two-page starter and build from there will get actual practice.
Discovery Education's "Education Insights 2025-2026: Fueling Learning Through Engagement" surveyed 1,398 superintendents, teachers, parents, and students. Ninety-three percent of educators surveyed agreed that student engagement is a critical metric for understanding overall achievement. Ninety-nine percent of superintendents called it one of the top predictors of success. Yet the survey also documented ongoing uncertainty — among all four groups — about what engagement actually looks like, how to measure it, and how to sustain it across classrooms. High agreement on importance. Low agreement on definition or evidence.
When a profession achieves near-universal agreement on a metric's importance while simultaneously failing to agree on what that metric measures, the metric has become a slogan. Engagement is doing that work right now in schools. A student making eye contact is not necessarily engaged. A student who answered incorrectly, reconsidered, and tried again is. The practical problem this survey surfaces is that without a shared, behavior-level definition, "engagement" is whatever the observer decides it is — which means observation tools, supervision conversations, and professional development all point in different directions while using the same word. Teachers who can describe engagement in terms of specific student behaviors — revision, self-correction, persistence after error — are operating at a precision level most of their evaluators are not.
A February 2026 policy brief from the American Institutes for Research examined state laws allowing students to take excused absences for mental health reasons — policies now on the books in approximately 10 states. The brief analyzed the relationship between these laws and chronic absenteeism rates, looking at whether formal permission to miss school for mental health reasons reduced the overall burden or complicated it. The findings are mixed: excused mental health absence policies have not demonstrably reduced chronic absenteeism rates, and in some cases have made attendance communication with families more ambiguous. Separately, RAND data from the 2024-25 school year shows that in roughly half of urban school districts, more than 30 percent of students missed 18 or more days.
Mental health day laws were designed with good intent, and the intent is not the problem. The problem is structural: a law that permits absences does not address why students stop coming to school in the first place. Chronic absence is rarely a single-cause phenomenon, and students who are missing 18 or more days are not missing them because they lacked a legal excuse. The AIR brief's most direct implication for classroom teachers is this — a student using mental health days at high frequency is sending a signal that the law is now making easier to categorize and harder to act on. An excused absence closes the attendance record. It does not open the conversation that needs to happen.