We hear a lot about how various industries are using data visualization and analytics. But what about the education industry? The institutional research office (IR) at universities is the center for data, reports and analytics and provides decision makers with information about the university. The IR teams are working on some 
innovative projects to tackle important issues within education. They're using predictive analytics for recruitment and to predict enrollment and retention, using analytics to determine strategies and interventions that can help students succeed, and much more.
At AIR Forum 2018, I had the opportunity to learn how several universities are using SAS. Here’s just a taste of what I learned:
The University of Oklahoma: Recruitment
The University of Oklahoma is using predictive analytics to assess the probability that an admitted student would enroll with the university, and determine what actions recruitment officers should take to entice students to enroll. The project was completed in five weeks and the models achieved 89-92 percent accuracy. By using predictive analytics, the university had its largest and most academically prepared student body ever.
George Mason University: Analytics in the Cloud
GMU shared how they went from having a variety of data sets disbursed across the institution to one unifying environment where they could put all of their data in a cloud-based infrastructure: the SAS Platform with Viya running in the cloud. They now have a single platform that can handle reporting, visualization, statistical and predictive analytics -- and machine learning.
University of Texas System: Innovating with analytics
The University of Texas System is using SAS in several innovative ways:
- Salary and wage information
seekUT is a free online tool and website that presents data on the earnings and educational graduate outcomes from University of Texas System Institutions. Students and their families can use seekUT to help them make informed decisions about their education and their financial future. seekUT shows the median earnings of UT graduates working full time — by degree level in 300 areas of study — 1, 5, and 10 years after graduation, alongside median student loan debt and much more. By partnering with the United States Census Bureau, UT was able to create a national-level earnings metric (annual earnings) for any graduate who has wage records from the 49 participating states. - Retention
The UT System needed to quantify the dollar value of a 1 percent increase in student retention rates. It's generally accepted that increasing retention rate is beneficial to the institution in terms of increased tuition and fees, state appropriations, reduced recruitment costs, etc. So, UT dug deeper and quantified all the revenues and expenses associated with a 1 percent increase in retention/graduation rates in a sample four-year public academic institution in Texas, and calculated the net profit to the institution and student using the time value of money (TVM) concept. - Dual credit
The UT System embarked on a mixed-methods research study to explore the relationship between dual credit and student success at UT System academic institutions. They wanted to get a better understanding of the relationship between students’ dual credit participation during high school and student outcomes once students matriculate to UT System academic institutions. - Student success
Using enrollment management data (demographic, academic, financial aid variables), national student clearinghouse data, and state unemployment insurance wage records, UT System conducted a series of cluster analyses to identify specific student risk profiles for first-time freshmen and transfer students. Using this cluster analysis, they created student risk profiles and use them to inform a student success framework.
University of Central Florida: Finding multiple uses for analytics
The University of Central Florida is using SAS in several innovative ways:
- Drop/Fail/Withdraw (DFW) rates
While analyzing DFW rates, UCF discovered that transfer students had higher DFW rates in all foundational courses examined than first-time college students in those same classes. Using this information UCF now has strategies and interventions planned and undertaken by the student affairs division in order to help students be more successful. - Budgeting
In another UCF project, executive leadership desired a new budget model. Budget personnel reached out to the IR department to provide the necessary details to support the new, complex structure. Subsequently, the IR department defined the problem and used SAS to develop a flexible solution that provided multiple delivery options. - Faculty reporting
Each term, the IR office within UCF receives data requests for faculty headcounts, tenure, promotion and retention. The old process for answering these requests required multiple steps, from extracting query results to creating reports. Now, using SAS, the process was reduced from five steps to one. This was done through the creation of a warehouse table with new fields added or calculated based on the most popular data requests, leading to a significant increase in efficiency within reporting processes. The IR process was greatly improved, which had a positive effect on dozens of data reporting tasks.
University of North Carolina at Greensboro: Enrollment
Given the increasingly robust collection and quality of university data, opportunities for predictive analytics are greater than ever. Accurate prediction of future student enrollment is vital for university planning efforts and decision making. UNC-G is using SAS to develop dynamic enrollment-projection models, with a focus on Markov chain modeling.
If you're interested in knowing more about how education institutions are using SAS, download this free ebook: Ten Tips for Using Data Visualization and Analytics Effectively in Education.
