Understanding Campus Life Through Digital Data

Date: 
November 9, 2015
Can aspects of a student’s life, such as mental health and academic performance, be predicted by evidence of their digital footprint as they live their daily lives? It’s a question Georgia Tech researchers want to explore with a new project.

The project is an extension of Dartmouth College’s StudentLife study.

During a recent GVU Center Brown Bag talk, Dartmouth College professor Andrew Campbell, who started the StudentLife project, asked, “Who are the students that are going to excel? Who are the students who are going to struggle? And who are the students that are going to drop out?”

WATCH: GVU Center Brown Bag Seminar Series - Dartmouth College’s Andrew Campbell discusses StudentLife study

Campbell says no one really knows because many factors impact academic performance and student life. StudentLife sought to answer these questions, though, using a continuous sensing app that assessed the day-to-day impact of workload on stress, sleep, activity, mood, sociability, mental health and academic performance of a small class of Dartmouth students over 10 weeks. Results from the StudentLife study show a number of significant correlations between smartphone data and mental health and educational outcomes of the student body. As the academic year progressed and workload increased, stress rose considerably, while positive affect, sleep, conversation and activity dropped off. Using this data, researchers accurately predicted student GPA, and in the future hope StudentLife will help students boost their academic performance while living a balanced life on campus.

Together with Dartmouth, Carnegie Mellon University and Cornell University, researchers at Georgia Tech are proposing to expand the study, now called CampusLife, by collecting data from a larger group of students through their interactions with mobile and wearable technology, social media, and the environment itself. In addition to providing valuable data for activity recognition in ubiquitous computing, the project has inspired researchers to think more boldly about a university campus as a testbed for understanding wellness.

“What drives this project is both a human goal of understanding wellness of young adults, as well as how one can perform such experimentation and address the significant security and privacy challenges,” said Georgia Tech College of Computing professor Gregory Abowd.

Abowd, along with College of Computing assistant professor Munmun de Choudhury, led a discussion about CampusLife as part of the IPaT Thursday Think Tank series. Faculty and staff from Georgia Tech and Emory also joined the brainstorming session.

The discussion focused on three questions regarding CampusLife: Who are the project stakeholders, what data should be collected, and with whom should researchers partner? Think Tank attendees agreed that the format of data collection is important; students may be more hesitant to provide information in an official capacity versus a more informal format such as social media. Offering value on top of data was also discussed. For example, going beyond data collection and using it to help students achieve work-life balance. Finally, the group talked about potential partners for the CampusLife project. One idea was to partner with health insurance companies as students reach the age where their parent’s insurance no longer covers them.

Researchers are working toward starting the CampusLife project during the 2017 academic year.