Nine in 10 colleges use some form of statistical analysis to determine retention and learning strategies.
Phil Ice knows numbers never lie.
Ice, vice president of research and development for American Public University System (APUS), has watched retention rates at the 97,000-student online school steadily climb with the continued analysis of in-depth information that shows when a student might be on the verge of dropping out.
If a student’s test scores are dropping, participation numbers are low, and disengagement is evident through various statistics, the numbers suggest that student might not last much longer at APUS.
How can professors and university officials know precisely which students are in danger of giving up on their education? One surefire strategy is to examine how many days have passed since the student last logged onto his or her course website.
If it’s been a while since the online student checked the site for syllabus updates or discussion sessions, APUS’s analytics system will flag the student as a potential dropout.
“We hone in on things such as a student’s perception of being able to build effective community,” Ice said.
APUS professors and instructors use information detailing a student’s online engagement with a comprehensive survey to create a model of student retention and satisfaction, according to the university.
“We’ve been using analytics before it was a buzz word in higher education,” Ice said.
APUS uses IBM’s SPSS Modeler, which uses key student performance, participation, and attendance information in part to measure a student’s social presence, along with a student’s perception of online learning’s effectiveness.
These two factors have proven to be reliable variables telling professors and instructors how likely a student is to drop out of classes.
APUS recently joined the Colorado Community College System, Rio Salado College, the University of Hawaii System, the University of Illinois-Springfield, and the University of Phoenix in a nationwide initiative aiming to improve analytics and increase its use in higher education.
The Predictive Analytics Reporting (PAR) Framework project, launched last May by the educational technology group Western Interstate Commission for Higher Education’s Cooperative for Educational Technologies (WCET), will examine six critical data sets as a single sample that could make analytics more effective on college campuses.
The six colleges and universities participating in the PAR project will create a pool of information from 400,000 students who will remain anonymous in the research.
Analysts will use this massive collection of student records to assess factors that affect retention and learning outcomes among online students.
Ice said about nine in 10 colleges use some form of statistical analysis to determine retention and learning strategies on campus, but only 9 percent use historical data to supplement current numbers—and only about 1 percent of U.S. colleges and universities use “deep data mining” to explore why dropout rates might be rising.
“That number really needs to expand, especially around the crisis of student enrollment,” Ice said. “We need to let the data speak for themselves.”
Campuses large and small have not been immune to spiking dropout rates.
University of Kansas officials worked with a data-mining company in 2010 to pinpoint strategies to keep students enrolled after a university report showed that 28.7 percent of freshmen from the fall 2007 semester have left the campus.
Kansas’s 28.7 percent dropout rate among fall 2007 freshmen was significantly higher than peer institutions, said Christopher Haufler, a University of Kansas professor who chaired the school’s student retention task force.
The university’s average retention rate after one year is 80 percent, he said, compared with 85 percent to 90 percent at peer institutions.
Kansas’s retention system uses analytics to raise real-time online flags for students who fall below a certain grade or involvement “threshold” designated by the university.
Nationwide, fewer than three-fourths of two-year career college students return to school after their first year, according to research released by the nonprofit Imagine America Foundation. Just 57 percent of public community college students return after one year, and 68 percent return after a year at a private institution, according to the research.
Other schools, such as Western Iowa Tech Community College (WITCC), use student tuition payment data to gage how likely a student is to leave school.
Educators often have a difficult time tracking the engagement level of online students, who—unlike traditional students—don’t interact with their professors and fellow students every day in class. Analytics tools, like the kind used at APUS, could make that task much easier, Ice said.
“By identifying patterns of performance, we believe we can create an approach to applying predictive analytics … that will help practitioners and students alike spot barriers to success before they become problems,” he said.