AI Spotlight: Using Predictive Analytics, Civitas Learning Helps Improve Graduation and Retention Rates for At-Risk Students

Share |

LISTEN: AI Spotlight Episode 3 is here.

It is no secret that the cost of postsecondary education is high. Obtaining a Bachelor’s degree can cost an individual
anywhere from $80,000 to $160,000 depending on the college or university – and that’s for students who graduate in four years. Unfortunately, less than half of students graduate in four years, and two more years of education only brings graduation rates up to 60%. For those students who do not finish their program of study, they leave an institution with no verifiable skills or credentials but have incurred significant debt – leaving them in a worse position to succeed than if they had chosen not to attend college. For universities, graduation rates matter too. Prospective students rely heavily on data like four-year graduation rates and year-to-year retention rates in their decision-making and state funding and continued participation in federal student financial aid programs for institutions depend on positive graduation and retention rates. 

In our artificial intelligence (AI) spotlight this week, we highlight Civitas Learning’s predictive analytics platform which provides higher education academic advisors and counselors with student data indicating which students are at-risk of not completing, need academic assistance, or where personalized outreach could improve further outcomes. Predictive analytics is the practice of extracting information from existing data sets to determine patterns and predict future outcomes and trends. Though predictive analytics capabilities have existed in the past, their use in artificial intelligence and platforms like Civitas Learning has been especially transformative in many aspects of life from health care to transportation. This platform demonstrates the transformative benefits of predictive analytics in postsecondary education as it has been proven to increase retention and graduation rates. Notably, Civitas Learning does not make decisions for students, but simply facilitates interaction between advisors and counselors, and at-risk students by identifying these at-risk students. In this way, the predictive analytics software supplements human work and judgement, encourages interaction between school administration and the students and helps students make the most out of their valuable educational experience.


Using predictive analytics to track student performance sounds complicated, but the idea behind Civitas Learning is fairly straightforward. Let’s say there’s a student who came in to college at the top of her class and thus is not labeled as at-risk, but in the first semester she struggles. She shrugs it off as a bad first semester, but this continues into the beginning of next semester. On top of the academics, she has to work full time to pay for school. Using data like her test scores and grades and comparing it to historical data of students from that same institution and program, Civitas is able to alert her advisor that she is beginning to slip off track and may require intervention to help her move back on track to success and credential completion. An advisor can then step in, meet with the student, and figure out exactly what may be happening in her life and studies and determine what the institution can do to ensure she moves back toward success. Interventions are then determined by academic advisors and could range simply from having regular conversations with the student to identifying courses she could potentially struggle in given her previous grades and develop a study or tutoring plan or suggest alternative courses. In doing so, Civitas improves outcomes for students tremendously.


In 2014, when the University of South Florida (USF) integrated Civitas’ platform into their management and information systems, the results were staggering. In just three years after its implementation, first-year retention rates had risen to 90% and graduation rates surpassed 70%, putting USF right up at the top with Florida’s strong public schools in these two categories. As a result of its performance, the university is expected to receive an additional $15 million in funding from the state. Beyond the data, the platform encourages counselors and advisors to reach out to students who are truly in need of help and creates a stronger, more impactful student-advisor relationship.


There are notable concerns with using predictive analytics in advising students. First, the data could be used inappropriately to weed out at-risk students to quickly increase graduation and retention statistics. Another concern is that students who struggle in one or two courses will simply be pushed out of the major rather than using such information to develop an academic plan for success in such courses. Thus, it is critical to have trained academic advisors and counselors to help students interpret their own data and decide how they should proceed that is in the best interest of the student. In this way, using predictive analytics should facilitate dialogue between students and advisors in order to improve student success and outcomes.


In SIIA’s Issue Brief on Artificial Intelligence and The Future of Work, we emphasize how AI is a natural outgrowth of the developments in computer technology like changes in data size, memory and processing speeds. Thus, these developments will lead to innovations not just in niche markets like higher education, but they will also have the ability to drive innovation across all levels of education, health care, transportation, speech recognition, and many other markets. Additionally, though some low-skilled jobs characterized by manual tasks may be replaced by AI, more job opportunities in patient care, construction, and high-skilled technical work are all also natural outgrowths of innovations in AI. The transformative benefits brought by AI in many aspects of daily life will continue to become more apparent and universal as we move towards a future defined by technological innovation in AI. 

Niko Nikola Marcich is an intern with the SIIA Policy team. He is currently an undergraduate student at the University of Virginia studying foreign affairs and international economics.