How Analytics Is Changing Student Retention and Success
Posted by Jami Morshed
Boosting student success is a key benchmark and those HEIs which can achieve it give themselves a major competitive advantage over their rivals. Identifying college students at risk is becoming a top priority for institutions as they look to devise early interventions to give them the best chance they can of graduating. The New York times recently published a story about a consortium of 11 large, public research universities established in 2014, with the goal of increasing retention and graduation rates among roughly 400,000 undergraduate students around the country. We will see more and more initiatives like this one as graduation rates become a top indicator of success.
By using research and hard evidence, rather than guesswork, administrators have powerful tools to effectively help students and increase their appeal to future applicants. Studying can be a captivating and exciting time. Equally, higher education is often the foundation of a career and personal development. But sometimes, institutions’ and students’ expectations do not coincide in the classroom, and students become disengaged and struggle to complete their degree program.
However, big data and analytics offer a solution to keep these students engaged and help to keep graduation rates high and appealing to potential students. By analyzing student success data, colleges and universities can tailor and refine their approach to defining degree requirements. Administrators can see hard evidence of what works and what doesn’t when it comes to retention and graduation rates. This allows universities to streamline academic journeys and boost student success rates.
What is “Big Data Analytics”?
Big data analytics is the aggregation of large amounts of data to make analytical decisions about future outcomes. IBM’s definition is “data sets whose size or type is beyond the ability of traditional relational databases to capture, manage and process the data with low-latency.” This can come from “sensors, devices, video/audio, networks, log files, transactional applications, web and social media - much of it generated in real time and in a very large scale.”
This approach can be applied to all kinds of industries. A variety of statistical techniques, such as regression analysis – the statistical modeling approach of defining the relationships between defined variables – are combined with statistical and spreadsheet software to achieve tangible solutions.
How colleges and universities have executed analytics
Higher education institutions (HEIs) use predictive analytics to take the guess work out of helping students succeed and graduate on time. Administrators look at the students’ characteristics – grades, retention and graduation rates – to track their journey through school. The analysis can prompt changes to degree requirements, aspects of student life, course syllabuses, grading procedures and more to ensure future students have a satisfactory learning experience.
Predictive analytics for student success project
In 2010, the University of Maryland University College (UMUC) paired with two local community colleges, Montgomery College and Prince George’s Community College, to run the Predictive Analytics for Student Success Project. Their objective was to mine data on general academic success at the schools, then observe how successfully students who transfer to UMUC make the transition. The report used first-term grade point average (GPA), retention, re-enrollment, graduation and course completion data to measure success.
By measuring course-taking behaviors at these schools and academic outcomes upon transition to UMUC, the schools could determine what makes students either successful or otherwise. They could then mine that data to better advise future students on how best to transition from community college to UMUC. The report came away with the following key findings:
- Students who took a math course at community college had higher correlated success in first-term GPA, re-enrollment and graduation.
- Success in a student’s first term at UMUC had a high correlation to re-enrollment and graduation rates.
- Online classroom engagement paired with a high GPA at a community college led to higher course completion in a student’s first semester at UMUC.
“Student Success Collaborative”
The University of Akron employed its own analytics study called the Student Success Collaborative. This measured student behaviors throughout their academic journey to gain insight into at-risk students, transfer behaviors and critical courses, to provide guidance and career preparation. Using this information, the University of Akron can identify graduation rates based on a certain GPA after the first year of school, or even based on a single grade in general courses, such as the introductory level classes of psychology and sociology. Administrators can also aggregate data to pinpoint which students will potentially transfer or drop out, and collect data on career path statistics. They are even collecting data on specific professions to see how people have landed that job. This will allow the school to advise on majors, minors and courses they should take to achieve their dream job.
Identify and support students
The benefits unlocked by the Maryland and Akron projects are not exclusively achievable through the development of proprietary technology. By designing the institution and supporting core systems around students, it should be possible to identify and intervene with at-risk students while there’s still time to turn them around. Include built-in Business Analytics, and institutions can gain a historical view of the impact of student success programs so they can be refined and improved over time.