Universities look toward data to stabilize enrollment numbers

Data mining could be one way Kansas officials address a decline in student retention.
Data mining could be one way Kansas officials address a decline in student retention.

University of Kansas officials are considering working with a data-mining company to pinpoint strategies to keep students enrolled after a recent report showed that 28.7 percent of freshmen from the fall 2007 semester have left the campus.

Five months after University Chancellor Bernadette Gray-Little formed a task force that would examine ways to increase student retention and graduation rates, task force members say they might enlist the help of Virginia-based data mining company Starfish Retention Solutions, which works with 14 four-year colleges, seven two-year campuses, and two K-12 school systems.

Starfish’s retention program helps campus decision makers weed out data that identify at-risk students with consistently low grades and spotty attendance records who are not engaged in campus activities.

The system connects to a campus’s course management system—Blackboard, for instance—and raises real-time online flags for students who fall below a certain grade or involvement “threshold” designated by the university, according to the Starfish web site. Faculty members can review these alerts through the school’s Starfish education technology portal.

Christopher Haufler, a University of Kansas professor who chaired the student retention task force, said the 20-person group examined several ways the school could bolster retention rates: incorporating more critical thinking and research into undergraduate courses, changing advising techniques, and intervening when students show telltale signs of academic slippage.

Haufler stressed that the university was not interested in tracking undergraduates’ every move on campus, but task force members saw data mining as a way to help students who fall behind on class assignments and drop out within a semester or two.

“We’re not going to become Big Brother and look at every move by the undergraduates,” Haufler said, adding that the task force report hasn’t been made public yet. “That, to me, is beyond where we ought to be. … But at the same time, we have a chance to monitor the progress of our students, and if a student hasn’t been going to class, then somebody should do something about it. Our [current] response time is slow, so we lose students before we ever get a chance to help them out.”

Kansas’s 28.7 percent dropout rate among fall 2007 freshmen was significantly higher than peer institutions, Haufler said. The university’s average retention rate after one year is 80 percent, he said, compared with 85 to 90 percent at peer schools.

The Starfish data-mining system caters to campus counselors as well as professors. The solution gives advisers and counselors contact information and student pictures, along with prior and upcoming advising appointments. Users also can see any of the student’s recent flags. The University of Kansas, Haufler said, has not committed to using the Starfish program, and it would use only certain features of the data-mining system if officials chose the system.

The task force examined other ways to improve retention rates, including the creation of smaller discussion groups in classes with hundreds of students. While expansive lecture halls are a central part of a large four-year university like Kansas, Haufler said professors should focus on forming smaller groups that encourage student questions and discussion.