Between dwindling enrollment rates and budgets tightening, the past year has increased stress on higher education institutions to not only meet enrollment goals, but hold onto the students they already have and ensure degree completion. In fact, every year, we see the “summer melt” [1] push nearly one-third of students off college campuses – and this year was even worse as students faced new challenges brought on by the pandemic.
The good news is that predictive analytics can help institutions in their fight to retain students, even amidst such tumultuous conditions. We’re already seeing the benefits of data analytics applied to student enrollment and retention, with some institutions reducing their summer melt attrition by approximately 1% in fall 2019 [2]. According to the “2020 Higher Ed Enrollment Trends Pulse Report” [3] by Othot, “institutions that actively used predictive and prescriptive analytics throughout the 2020 enrollment cycle countered the national trend of steep enrollment declines.”
Predictive analytics focuses on the future while prescriptive analytics determines immediate actions that institutions can take to improve enrollment and retention. Using modern data science methods, institutions can implement both predictive and prescriptive analytics to define how student and prospect marketing campaigns are performing and make suggestions for changes that will create better student reception and action.
For example, Florida Institute of Technology increased enrollment rates in 2020 by 3% compared to 2019 enrollment by using predictive and prescriptive analytics. Another example is Texas Tech University, which grew its enrollment to 40,322 in 2020, 322 more than its original enrollment goal of 40,000 – with the largest freshman class in the university’s history. On top of that, Texas Tech increased enrollment by a whopping 9% and student retention by 2.6% across a three-year span simply by using predictive and prescriptive analytics.
In Texas Tech’s case, predictive and prescriptive analytics came into play to determine students with a high likelihood of enrolling and persisting and how marketing efforts would influence those outcomes, e.g., which students out of 300,000 would get a mailer. Ultimately, Texas Tech was able to increase the prospect pipeline while working with a flat budget in 2020 by being more strategic through the use of analytics.
Taylor University, a Christian liberal arts university in Indiana and ranked as the No. 2 college in the Midwest region, is also looking to the power of analytics to meet its enrollment goals. Using advanced analytics and data science methods, the institution set out to increase conversion rates in every stage of the enrollment process to fight competition around prospective students and enroll individuals who were a better fit and more likely to complete their degrees at Taylor University. This would ultimately allow the university to increase tuition revenue, even without growing headcount because it meant they were enrolling students who would actually stick around.
Using a combination of around 60 data points including metrics such as GPA, academic interests, test scores, engagement and more, Taylor University’s enterprise data systems team harnesses predictive analytics to determine which variables have the highest predictive strength in each step of the enrollment process. As admissions recruiters review an application pool of more than 50,000 potential students, predictive analytics helps them identify students with the best fit for the university and highest likelihood of retention to fill the institution’s 500 open spots. With a combination of predictive analytics and a proactive recruitment strategy, Taylor University has seen record growth.
After only one year of implementing predictive analytics into its enrollment cycle, Taylor University enrolled the largest freshman class in its history. A year after that, the university again recruited a record freshman class, which came in as the fourth-largest class in school history.
When it comes to community colleges, these institutions face a completely different challenge. On top of the hard work behind recruiting and enrolling new students, community colleges face a higher rate of attrition than traditional institutions. Why? Many students attending these types of institutions face pressure from outside factors such as full-time jobs, family obligations, transportation challenges and more. The focus here is on retention. Predictive analytics can guide intervention strategies for struggling students at community colleges.
According to Brian Merritt, chief academic officer and vice president for learning and workforce development at Central Carolina Community College (CCCC), the past year has meant many students lost income and are stressed from the myriad pressures facing them, leading to the greatest student drop in nearly a decade for community colleges. However, he says, “the more nimble we can be to make needed adjustments to our policies and practices, the more likely we can keep their momentum going through this pandemic.”
CCCC is using predictive analytics to do just that. Gathering data from data sources such as the student information system, learning management system, enterprise resource planning system and more, predictive analytics allows the institution to make better predictions around students’ potential for degree completion and retention. It also allows the academic advisors and success coaches to have better, more engaging interactions with students and prompts them to intervene when red flags arise.
Community colleges also face the same pressures around recruitment and enrollment as traditional institutions. CCCC is facing these challenges head-on by using predictive analytics to give recruiting, marketing and enrollment staff direction for where to focus efforts based on geodemographic data and other information.
Using predictive analytics, CCCC has increased retention by 9% for full-time students and 18% for part-time students on average since 2012. Increased retention has also yielded higher graduation rates for the institution, with a 19% increase in students completing their degrees – that’s thousands of students.
Predictive analytics and prescriptive analytics have the potential to make a huge difference for higher education institutions struggling to enroll and retain students, and for students struggling to stay engaged. By enlisting the power of technology, higher education institutions can make a difference in millions of students’ lives by helping them fight for their future.
Bryan Bell was the Chief Data Scientist for Watermark (formerly Aviso Retention) where he was responsible for analyzing data and ensuring that Watermark’s efforts are adequately aligned with the needs of students and education leaders. A solutions-driven data scientist and entrepreneur with a passion for data storytelling, Bryan used data science to support the overall mission of Watermark.
This article was originally published in Informs Today on 6/22/2021
