Access the best custom research to help hit your organization's goals. Set up your custom consult below in 60 seconds or less
to general questions, press inquiries, references, citations & more. Just fill out the form below and we.ll get back to you within 48 hours or less
Colleges and universities squeezed by budget cuts stabilize revenue streams by improving student retention. The surprising connections these institutions uncovered through data mining and predictive models have informed improvements to student retention strategies:
Through data mining and analysis, administrators at South Texas College discovered that students who register late for a course are more likely to fail or withdraw. Realizing that this has a negative impact on student retention and time to completion, South Texas decided to eliminate late registration.
Students in a graduate course built a predictive model of student retention using SAS software. One of the group’s findings was that commuter students are more likely to drop out of the university. Consequently, the university developed student retention strategies including requiring all freshmen to live on-campus and recruiting at-risk students to participate in specialized seminars and other programs.
“Signals” is a personalized program run through Blackboard. According to the Signals website: “To identify students at risk academically, Signals combines predictive modeling with data-mining from Blackboard Vista.” After the data are collected and analyzed, students are assigned to a specific retention risk category. Students see this category on their Blackboard course page. They see a green “signal” if they are in the low-risk group, a yellow “signal” if they are on the edge, and a red “signal” if they are at risk of failing. Based on this information, professors can send out customizable emails to students in each risk group. This student retention strategy will help point students to important academic resources.
The engineering school at SUNY-Buffalo rates incoming students on seven academic factors related to their preparation and test scores. If a student is substandard on five of these factors, then the school advises the student to enroll in specialized tutoring sessions for entry-level courses.
Through predictive modeling, Tiffin found that academic, financial, and social factors all affected retention risk. The university’s new student retention strategies include assigning high-risk students to personal mentors, designating a chief retention office to monitor students in the medium-risk group and sending an automated email message to students in the low-risk group to inform them of relevant extracurricular activities. Through these efforts, Tiffin improved its student retention strategies: Tiffin’s one-year student retention rate increased from 51 percent to 63 percent in just five years.
MAP-Works is a comprehensive data-mining tool developed by Educational Benchmarking (EBI) and Ball State University. MAP-Works combines institutional data with student survey responses to provide a complete retention risk analysis of first-year students. First-year students are asked to complete a 15-minute “Transition Survey” the third week of the semester. Questions cover academic, social, and emotional areas. Survey responses are available to faculty and staff, who then suggest appropriate interventions and new student retention strategies. Students are also able to compare their survey results to aggregated student responses.
According to EBI’s website, six postsecondary institutions use MAP-Works: Slippery Rock University, University of Illinois College of Business Administration, Ball State University, Hastings College, Iowa State University, and Casper College. These institutions reported the following results:
Data mining and analysis helped these 11 institutions improve student retention strategies, through simple changes to registration and housing requirements as well as through larger efforts to monitor and support at-risk students. Many colleges and universities collect student data; better research will help decision-makers utilize it.