Educators have been struggling for decades to resolve a fundamental problem: Students who are in the same grade because of age often vary greatly in skills, abilities and experiences, even on the first day of kindergarten.
Teachers are told to differentiate their instruction so that each student gets what she needs — a good idea in theory, but hard to pull off in a real classroom because teachers also vary in skills and abilities.
That’s the big puzzle that University of Washington computer science professor Zoran Popović hopes to solve with insights gained over the last five years of developing computer learning games that adapt to the skills of individual players so they progress more efficiently toward mastery.
Popović directs the university’s Center for Game Science.
He also is the founder and chief scientist at Enlearn, a not-for-profit organization started with money from the Bill & Melinda Gates Foundation, which partnered with the center in May. Enlearn is developing a commercial application for the interactive technology aimed at the global K-12 market.
Enlearn has created a single computer platform that can digitize a standard math curriculum and then customize specific problems to an individual student’s needs, based on computer algorithms developed for the learning games.
As students work through the problems on tablet computers, the data is fed to the teacher’s tablet, providing a moment-by-moment progress report on how each student is faring and whether the class as a whole is ready to move on or needs a better explanation.
“The platform, in real time, provides the key misunderstandings and misconceptions for every individual student, which directly informs the teacher about what to do next at that instant,” Popović said.
In other words, turbocharged differentiation.
This spring, Enlearn tested the platform in nine sixth-grade classrooms within three schools in the Seattle school district.
The 5-day trials focused on ratios and proportional reasoning using two versions of a standard math curriculum: the standard paper one and a tablet version powered by Enlearn’s platform.
Students took pre-tests that showed some math classes had stronger students than others. So Popović assigned the tablets to the weaker classes, essentially stacking the deck against the Enlearn version.
Nevertheless, the students with the tablets solved 4.5 times more problems on average than the kids in the stronger classes using the paper version, according to Enlearn. Teachers tried it both ways, but they were able to assist three times as many students in the same amount of time in classrooms using the tablet version.
The results are being prepared for submission to a peer-reviewed journal in September.