In New York City Public Schools’ District 22, teachers used forward-looking insights to drive stronger student outcomes in just three months.
After three months, one classroom saw an average +32 point increase across different standards. Pilot teachers using the predictive feature saw an average of +36 points increase from their students' predicted performance vs. their actual performance.
District 22 was in the middle of rolling out Illustrative Mathematics (IM), with schools at different stages of implementation. District leadership consistently heard, especially at the elementary level, that teachers needed more than a new math HQIM. They needed support to teach it well. Their superintendent heard more about HQIA, and realized how essential assessments are for connecting instruction and curriculum.
That was the backdrop when the District was introduced to Compass Math. For a district juggling a multi-year curriculum transition, the promise of curriculum-compatible support that built teachers’ confidence in their materials resonated.
One participating fifth-grade team used its double math block, extra time already built in for intervention, to put Compass Math data to work. Teachers used it to plan “WIN stations” (“What I Need” groupings), rotating students through small groups based on specific misconceptions.
That level of detail changed the size of the problem teachers were solving. Instead of reteaching an entire concept like adding and subtracting fractions, they could isolate the exact sub-skill tripping students up — finding a common denominator, say — and pull a small group accordingly, backtracking to earlier-grade content when a gap traced further back. Because Compass is built around the same curriculum the school already teaches, teachers found they rarely needed to look elsewhere for materials to close those gaps.
For school Leader Dr. Vlantis, the value of Compass Math showed up in the conversations it enabled with teachers. Screener data tends to surface the big picture, with broad concepts where a student might be struggling. Compass Math went further, using answer distractor rationale based on 2.5 million student data points to reveal why a student answered the way they did.
But getting there took a mindset shift. Early on, the instinct was to treat Compass Math data as a backward-looking record of what students had already gotten wrong, like most assessment data. The professional learning built into Compass helped reset that instinct.
“When we initially gave milestones, it was like, 'Wait, if I'm looking at this for backwards data, how is that helping me guide instruction?' Once we sat in PD with Compass, you realized, 'No, we have to look at these predicted strengths and here's how to use it.'"
That pattern went beyond a single classroom. Across District 22's pilot teachers, the trend continued. In several cases, the actual scores more than tripled the original projection.
Consistent with how the pilot began, its growth is being driven by schools. PS 277 is expanding Compass Math from its original pilot into grades 2–5 for the 2026–27 school year, and two additional pilot schools have confirmed they'll continue using Compass Math this coming school year. District leadership remains enthusiastic about what the pilot has shown and what this means for their students.
Compass Math closes the gap between data and action. Built on two decades of assessment experience and a nationwide dataset of student misconceptions, Compass Math gives you insights you can actually use when they matter most. Explore how Compass Math can work in your district today.