Higher education institutions are being challenged to balance their mission as educators and their need to support that mission financially. They are also seeing more scrutiny around justifying the value of an education, which can have far reaching implications. Today, many states are reconsidering how they fund higher education and are seeking to link funding to key performance indicators like student retention and graduation rates. “Experts” and media pundits are speculating whether this will lead to ‘winners and losers’ in this new model.
That's changing the way higher education institutions operate, and many are using data and analytics to meet these new demands.
Our vision at Othot is to bring advanced analytics solutions to critical stages in the relationship between student and institution that will improve decision making. Our early work has been focused primarily around the enrollment cycle; however, we have learned a tremendous amount about student behavior along the way and that’s formed a great foundation for us to approach student retention.
For example, we have created highly predictive models around questions like what is the likelihood to persist to sophomore year? during the enrollment cycle as an additional prediction that is incorporated into our enrollment analytics. Current customers have seen value in either looking at the persistence score as an input into their enrollment decision or, later on in the cycle, as a valuable input into shaping the class, mitigating summer melt and helping at-risk students.
Because we develop dynamic predictions at the individual level, we can also help customers look at the deposit stage in the enrollment cycle to identify students that are most at-risk to melt.
Finally, we have worked with some select customers to create models that focus on persistence through sophomore year before classes even start. Customers have identified students who are most at-risk and used this information to create and implement mitigation strategies and tactics before triggering events have even occurred.
A recent study found that 34% of students who did not return after freshman year had a 2.0 or lower. Root causes for why a student doesn’t graduate may include academic issues, lack of support, finances, family or general immaturity1.
With Othot’s open data architecture, we pull in data from multiple data sources and in a dynamic way, which is oftentimes a huge hurdle for creating models that are more multi-dimensional. Our machine learning algorithms allow for likelihood scores that contemplate the impact that multiple variables have on an outcome. Finally, the ‘what-if’ functionality in our platform allows users to explore what prescriptive variables might help improve a score.
We also value agility and speed. Our cross-functional team has designed our onboarding process in such a way that we are getting customers onto our platform in less than two months. Time to value is critical.
If you are actively considering how to develop a more robust retention strategy or are even looking to tweak or refine what is already in place, we’d love to help you. Our team would be happy to share a product demo with you to show you how it works. Please contact us at email@example.com.
1 Civitas Learning, Emerging Benchmarks from Across the Civitas, October 2017
By Dave Babst | April 19, 2018