Ball State University – Boosting Retention with Predictive Analytics
In 2006, Ball State University was faced with a growing problem. Over a five year period they saw their retention levels decline below 75% and they had an enrollment decline of 700 students. They knew it was time to take action.
Ball State believes that every employee is responsible for student retention. The Student Affairs division took the lead in heading up their new retention initiatives with Dr. Kay Bales, the Vice President of Student Affairs and Dean of Students heading up the efforts.
At first, the team focused on things they could accomplish in the short term- additional feedback loops based on results of student surveys, and other outreach programs to make sure they were connecting with the right students. A Retention & Graduation Specialist was added to the staff in 2009 and another in 2012 as things continued to improve.
They began to gather data: a student’s academic preparedness from admissions, card swipe data from dining, recreation services and other student life engagement data, financial aid data, and survey data “Like Hansel and Gretel, students leave a digital footprint on campus. They provide information to us about their engagement on campus. With the right tools we thought we could follow their digital footprint.” said Dr. Bales.
As they gathered all of this data, they had challenges on how to combine all this siloed information and easily determine which data points were most useful. Some students would appear on multiple lists but, the university had no way of knowing which variables were statistically significant. They needed to find strong analytics and data preparation tools to help them make sense of all of the data. They looked to Rapid Insight Veera™ and Analytics to help them address their issues.
With a freshman class of 3600, they were struggling to effectively identify which 700 to 800 students to provide outreach to who were most likely preparing to leave the university. As Ball State began to use Rapid Insight software, they were able to go beyond some of their standard assumptions and integrate an additional 250+ variables in their analysis- resulting in more precision. The list was reduced down to 400 students who had a 50 percent chance of not finishing.
“No student outreach is a waste of time. But with the Rapid Insight tools our team now has more confidence that we are accurately pinpointing the students who will benefit most. “said Dr. Bales.
Dr. Bales team has even been able to quickly test some hunches they had- discovering a correlation that students who eat breakfast in the morning are more likely to persist. They continue to refine their models with this new information so the models are never outdated. It has created a mindset change on campus, allowing them to test out theories before they begin new retention programs and be better stewards of their limited resources.
“By bringing our predictive modeling and analytics efforts in-house, we are able to continually update the model so we aren’t building strategies on data that is three months old. We are working faster and smarter reaching the right students.“ continued Dr. Bales.
Through all of their campus-wide initiatives, Ball State has improved their retention rate by over 6 points since its low of 75%. Ball State is also experiencing the largest five-year increase in on-time graduation rates of any public institution of higher education in the state of Indiana.
As Ball State looks to the future, they know they have only scratched the surface of the potential insight they can glean from analyzing their retention and student behavior data as they continue to evolve as a data-driven institution. Dr. Bales and her team are going to be introducing over 100 more variables into their retention model to see the impact it may have. In addition, they are partnering more closely with their enrollment services division, who also utilize the Rapid Insight solution, to see how the partnership can further strengthen their retention model.
“We really do see the possibilities as limitless as we do everything we can to be more student-centered.“ noted Dr. Bales.
The leadership of Dickinson College understands the importance of predictive analytics to help shape their class. But they also know that models can get stale, and will only improve when they are transparent and institutional departments collaborate during the model...read more
Clark University initially became interested in predictive modeling due to an 87% growth in applications within three years. As exciting and encouraging as this was, it also presented Clark with a new set of challenges. In order to acclimate to the rise in student...read more