NOV 10, 2026 · DOANNIE TRAN, MANAGING PARTNER, CENTER FOR INNOVATION IN EDUCATION

Can we have an assessment and accountability system that makes teaching and learning better?

In New York, a network of 38 public high schools have replaced state standardized assessments with performance-based tasks–without widening educational inequity or overburdening their educators. In fact, graduates of New York Performance Standards Consortium schools earn higher first-semester college GPAs and persist in college at higher rates than their peers while these schools enroll more English language learners (21% vs. 14% citywide), more students with IEPs (23% vs. 19%), and more students who arrived in 9th grade behind on traditional measures.

This calls into question two things we have always been told about assessment and accountability:

  • Accountability requires identical tests. Only the same tasks under the same conditions yield data that we can meaningfully compare.
  • Teachers can’t handle much more than the current system. A system based on real student work would be too much for an already over-taxed workforce.

As a part of the Educational Accountability Remade project, we looked at systems that build community trust and support educators. In two new papers (Comparability Without the Harms of Standardization and Authentic Assessment and Accountability as an Engine of Educator Renewal) we argue that both of the above beliefs rely on assumptions worth challenging.

Can we get comparability without uniformity?

Today’s accountability system relies on data comparability. Standardize the tests and how they are administered and you can compare the scores across contexts. But that creates a testing environment that goes against what we know about learning. Students need context, relevance, and connection to one another and their communities, but an accountability system built on decontextualized, individualized tasks causes teaching and learning to become irrelevant and isolating.

Contrast this to New York’s Consortium schools – and others like them nationwide – that are holding themselves accountable for deeper and more meaningful learning demonstrated through performance-based assessments. While these systems intentionally allow for cultural and contextual variation, they can still achieve comparability through a different approach described by psychometrician Robert Mislevy and his colleagues: construct comparability. Tasks can differ as long as they measure the same knowledge or skill and are judged against the same definition of mastery.

Construct comparability requires two things: 1. A shared definition of the skill through a rubric that describes quality criteria and 2. A clear argument for how each local task calls for that level of thinking. This approach depends on teachers scoring work together until they agree, which is described in greater detail in the full paper.

Read the Comparability paper (PDF) →

Can we have an accountability system that actually makes teachers better?

Concerns about teacher capacity to deliver on a more ambitious system are real. Designing tasks and scoring student work takes time and resources, and we describe what it takes in our paper. But the question of “can teachers do it?” assumes that the capacity to do the system must preexist before we can change. We make the case that the capacity to run a meaningful assessment and accountability system grows through the work of building the system itself.

In the New York Consortium, teachers score student work together twice a year against shared rubrics, building their own carefully calibrated professional judgment across schools in evaluating student work against agreed upon standards. Those teachers report greater influence over their work than other NY teachers (83% vs 68% positive) and are more likely to have stayed at their school for three or more years (86% vs 78%). These are associations, not necessarily causal relationships, and we acknowledge that teachers that value this kind of work may preferentially choose these schools. However, we also know that the current system of standardized assessment and accountability is causally linked to increased educator stress, burnout, and turnover. We need a different system, one where investing in assessment and accountability also invests in educator capacity and renewal.

Read the paper on authentic assessment and educator renewal (PDF) →

What this means for your community

If comparable doesn’t have to mean identical, and capacity can be built through the work, then your state, district or network could design a different kind of system. We want to hear more about what you’re thinking and trying – find us at www.eduaccountability.org.

← All posts