Dartmouth students can predict their GPA with their smartphone

StudentLife uses data collected from smartphone sensors to monitor students' habits, predicting their GPA score within 0.17 of a point.

Any student who has partied that bit too much will know the impact it can have on their GPA. But beyond late nights, there are a whole range of cognitive, behavioral and environmental factors that can impact academic performance — including quantity and quality of sleep and amount of physical activity. Now, researchers at Dartmouth College and the University of Texas at Austin are using a SmartGPA app to carry out their StudentLife study, which uses data collected from smartphone sensors to monitor students’ habits and predict their GPA score to within 0.17 of a point.

The SmartGPA app tracks numerous behaviors — such as length and focus of study time, frequency of partying, sleep habits, class attendance and exercise — without any direct input from the student. Instead, it uses information gathered from their smartphone and uses machine learning algorithms to produce a prediction of their GPA without any prior knowledge of the student’s SAT scores or IQ. The app was tested on 30 Dartmouth undergraduates over a 10-week term and predicted their academic performance with impressive accuracy.

SmartGPA is an excellent tool for helping students get the balance right during college. We have seen other apps such as Pocket Points which encourage similar self-restraint. Are there other ways to use smartphones as a stimulation instead of a distraction?