Better analytics for new courses
Acrobatiq uses observations of student learning in a course — specifically, their selections and responses on questions and practice activities — to estimate progress toward course objectives and to predict outcomes on tests. Prior to this release, Acrobatiq used data from students in the first version of a course to improve its predictions for students in subsequent terms.
Our new analytic model starts learning about students while the first version of a course is running. After 25 students have gone through an objective in a course, the model learns from their responses, and tunes itself so that it can make better predictions about the next 25 students to work on that objective. Then, after 50 students complete the objective, the model tunes itself again. The more data the model gets, the better its learning estimates are.