For more than thirty years, adaptive learning and adaptive testing systems have been heralded as breakthrough innovations in education. From early pioneers like Carnegie Learning’s Cognitive Tutor™ in the late 1990s to today’s i-Ready™, Mathia™, ALEKS™, and NWEA MAP Growth™, adaptivity has been marketed as the ultimate solution: a faster, more precise measurement of student ability, delivered through personalized pathways that “meet students where they are.”
And yet, despite billions in investment and decades of widespread adoption, student achievement in the United States has remained flat— and even declined. The most recent post-COVID NAEP results showed the steepest drops in reading and math in decades. If adaptive technology really delivered on its promise, we should have seen some positive signal in the data by now. We haven’t.
Teachers know why. Adaptive platforms may be efficient at measurement, but they are rarely transparent, actionable, or connected to the real work of the classroom. In contrast, modern formative assessment platforms, such as Formative.com™ or Classwork.com™, support data-driven instruction. This instructional approach is what adaptivity has always promised but never truly delivered: an engine for increasing student achievement.
Let’s break down the contrast, drawing from the classroom perspective, the lessons learned during COVID, and thirty years of experience with computer-centric models.
