University of Pittsburgh: Data-Driven Management for Enrollment and Retention
Seidel's enrollment model predicted within 1% of the actual class size
2 data-driven reports are distributed for assistance in enrollment planning
Now that student deposit predictions are underway, retention models are up next
The Office of Admissions and Financial Aid at the University of Pittsburgh was looking to improve data-driven management for enrollment and retention efforts. After researching other options, the university chose to implement the Rapid Insight platform.
Build Confidence in the Model
Michael Seidel, a Senior Data Analyst in the Office of Admissions and Financial Aid, began using the Rapid Insight platform with historical data to see the accuracy of a prediction. Seidel created an enrollment model based on Pitt’s most recent freshman class. The model predicted within 1% of the actual class size. Although new to the predictive modeling game, Seidel had a very positive experience with his first effort and was really excited about the accuracy of the data-driven management model.
Manage Enrollment and Admissions
Enrollment accuracy is essential to the survival of any university. Under-enrollment can lead to unfavorable financial complications and over-enrollment can lead to overcrowded housing and a shortage of available classes.
To stay on top of their enrollment, Seidel now runs two data-driven models a week. The results of these models are sent to the Director of Operations and the Chief Enrollment Officer and are used in decisions around enrollment planning. These reports consist of both general numerical data and specific student names; in case they want additional individualized outreach.
The weekly reports make predictions about whether or not a prospective student would pay their deposit. Seidel tested this method by creating a file that contained actual numbers from the previous year. The university used comparative measures to get results. They looked at students who had deposited in the past, compared to similar students. One data pool consisted of applicants, while the other consisted of current students. The institution was excited to see that they could predict results for the upcoming fall term applicants.
Support Student Success Efforts
In addition to using the software for student deposit predictions, Pitt has begun to focus on retention rates. By predicting student success rates, the university can determine the extent to which they need to reach out and support their students to ensure retention.
Pitt identified students from last year’s freshman class and from two years’ prior who performed lower than predicted. Seidel used Predict to sort the data into deciles, or bins of ten. Integrating the data with information from The College Board, he was able to detect the top two bins of students who looked most like the students who were identified as having underperformed during their first two or four semesters. It was discovered that 406 students were 29% more likely to not succeed. This prediction is based on their performance before college and within the first two years at the university. With grade reporting this December, Seidel wants to identify freshmen at risk and determine how close the numbers are. After that, he will work with the Chief Enrollment Officer to formulate a plan of action.
Within this data-driven management system, Seidel can pinpoint recurring trends and traits. For instance, he identified specific schools within the university that have a lower success factor. Results can often time be ostensibly intuitive, but sometimes yield interesting and unforeseen surprises. For example, students with the highest probability of not succeeding are those with SAT scores between 1250-1290 and 1350-1390.
“This is going to be really valuable moving forward. When you’re making admissions decisions, you want to look at this chart and ask, ‘Are they falling into one or more of these buckets?’ That might be a red flag,” Seidel said.
Data-Driven Management Future
Pitt plans to use Rapid Insight tools for prudent financial aid judgments. Seidel hopes to use predictive modeling to accurately and cost-effectively distribute financial aid funds. One of the most asked questions is: How much institutional money are we going to spend if these students enroll at Pitt? Rapid Insight’s data-driven management tools can answer that with Pitt’s existing data.
“Any time I’m looking to do analysis, I always try and use the software because it makes it so easy,” Seidel reflected. With the ease of data-driven management software, Pitt is eager to see where the data will take them.