Freshman Dropout Predictions

Have you ever wondered if spending more time in the library actually equates to better academic performance? The University of Arizona is tracking freshman students’ ID card swipes to anticipate which students are more likely to drop out.

The new ID card tracking system keeps a record of how often students interact in social settings on campus (like use the campus rec center), what they buy to eat, and their academic performance. According to the University, the data allows them to predict within a freshman’s first 4 weeks if they will return as a sophomore and eventually graduate.

Based on the data, the university identifies a list of freshman in danger of dropping out and shares it with the students’ advisors every quarter, who do their best to intervene. According to the article, students with shrinking social circles and a lack of a routine might be more likely to drop out.

The efforts have been pretty successful so far. After three years of collecting freshman data, their predictions have been 73% accurate. Last year, the school’s retention rate rose to 86.5% (almost 10% above the national average).

“We think by doing these interventions by the 12th week, which is when students make up their mind, you’re sort of doing what Amazon does—delivering items you didn’t order but will be ordering in the future,” says Sudha Ram, a professor of management information systems who directs the initiative.

Like any predictive technology, some major ethical concerns about privacy arose. It could be argued that this level of analyzing students’ social interaction data, which includes timestamps and locations, potentially violates students’ privacy. Still, algorithms can sometimes be wrong and biased. Ram admits, “We live in an era where you shouldn’t be generalizing about ‘groups of people. You should be personalizing solutions at the individual level.” She calls the data she’s analyzed “just a signal.”

Why It’s Hot: This initiative is using predictive technology in a much more meaningful way than say, suggesting what products you might also like to buy on Amazon. If this machine learning tool can identify behaviors that may lead a student to drop out, who’s to say it couldn’t be developed further to signify behaviors that lead students to attempt suicide or fall into depression? If possible, many students could receive help from advisors or family members who were prompted by the system.

Source: https://www.theverge.com/2018/3/12/17109224/university-of-arizona-tracking-smartchips-student-ids-privacy-drop-out