Skip To Content
Rapid Insight: Higher Education case study

Western Carolina Utilizes Rapid Insight Tools for Enrollment Forecasting

Rapid Insight tools

Western Carolina University faced a challenge that some institutions would not mind having – growing enrollment numbers.

While this excites Operations Research Analyst Johnny Lail, he also knows that predicting accurate enrollment numbers is important when they request funding dollars from the government. “Our enrollment projection numbers are what our state funding is based on. If we underestimate, we have left money on the table. If we overestimate, we owe money back,” he said.

When Johnny took the position at Western Carolina University, one of the first things that Johnny was tasked with was to help forecast enrollment for the incoming freshman class. Before taking his position, Western Carolina University had purchased the Rapid Insight platform, including QuickStart for Banner, a series of predictive modeling templates for enrollment management and student retention. Both templates are designed specifically for users of the Banner student information system. When Johnny came on board, he needed to learn the Banner database structure as well as the Rapid Insight products – and he needed to get up to speed quickly.

“The drag and drop interface within Construct, combined with the fact that the Banner data was already mapped out, made this a much easier process,” noted Johnny.

When it was time to work on his first enrollment projections, Johnny realized that he needed to factor in a twist into his projections that some schools may not encounter – the “Summer Bridge” program at Western Carolina. Bridge programs are designed to improve the preparation and ease the transition into college in the Fall. At Western Carolina, when a student applies and they do not meet the criteria to be part of the fall cohort, they may have the option of applying for the Summer Bridge program. Summer Bridge students are different than the freshmen who will start in the Fall, posing a unique modeling challenge. To complicate the effort further, many accepted “Fall” freshmen choose to start early in the summer.

The Rapid Insight QuickStart templates made it easier for Johnny to build predictive models for enrollment management and student retention/student success. In addition to providing a proven method for mapping, integrating, and organizing Banner data to create the analytic dataset, QuickStart customers have the option to customize these templates to meet their own specific needs. When Western Carolina was looking to incorporate the Summer Bridge students into their QuickStart templates, they contacted Rapid Insight to review the changes.

“Rapid Insight helped to make the QuickStart templates work for our individual needs. After a few back and forths, we came up with a solution that incorporated our unique needs with the research already done on the predictive variables within our Banner system,” Johnny said.

To test out the accuracy of the models, Johnny used three years of historical data to test 2012 and an additional year’s worth to test 2013. The projections were spot on, with 99.5% accuracy in 2012 and 99% accuracy for 2013. The big test came with the incoming 2014 class. Projections within the school were for an incoming freshman class of 1,650 students. The model was projecting 1,766 students, which was over 100 more than the standard campus estimates. When enrollment was finalized, the actual number ended up being 1,745 students.

“So basically, we had one of the highest jumps in the freshman class that we have seen in a long time. No one expected it, yet Rapid Insight predicted it,” noted Johnny.

In the long term, Johnny would like to be able to project his enrollment for all classes, including graduate students and transfers. This supports the vision statement of Western Carolina, “To be a national model for student learning and engagement that embraces its responsibilities as a regionally engaged university.” Johnny knows it will take some research on what the predictive variables will be. But, like the Summer Bridge twist in their Freshman Enrollment model, they know they will be able to get there.