As colleges and universities face mounting pressure to raise completion rates, many are embracing predictive analytics as a way to identify which students are at risk of failing courses or dropping out. But how accurate are those predictions? 

Data experts put six predictive models to the test.

Photo: Eli Imadali 

Like many high school students who graduated during the coronavirus pandemic, QuangHuy Bui, 19, didn’t go straight to college. After watching YouTube videos, Bui went to work as a barber. But the last year also clarified his bigger goals. He wants to own a barbershop someday—maybe even a franchise.

Driven by that larger goal—and his family’s refusal to let him get complacent—Bui is headed to the University of Colorado Boulder this fall, where he plans to major in business. He’s part of an influx of students picking up college plans that were delayed by the pandemic.