Throughout the undergraduate recruitment process, schools have the opportunity to capture huge volumes of data about student interests, behaviors and preferences.
Even a single university could plausibly collect millions of data points in the course of its search, application and yield campaigns. Here at EAB Enrollment Services (formely Royall & Company), by virtue of working with more than 300 institutions annually, we maintain an exponentially larger storehouse of real-world data than any single school would have on its own.
Over time, this data has helped us become more effective at engaging students on behalf of our partner institutions. It’s given us important inputs for yield models that predict whether admitted students will actually enroll.
So a number of our clients have been asking: Could this wealth of information also help schools predict which students are most in need of support in their first weeks and months on campus?
This particular application of pre-enrollment data was especially intriguing to us because student success analytics (such as those used in EAB’s Student Success Collaborative platform) can help institutions manage at-risk students more efficiently and effectively than ever before, by identifying indicators of risk from various on-campus data sources.
To date, however, these analytics systems have demonstrated a limited ability to identify students who are retention risks early in their college careers, simply because schools don’t have much predictive data about students who have just shown up on campus.
A challenge for our data scientists
This seemed like a perfect challenge for our joint EAB-Royall & Company data science team, who spent a year conducting research with six Royall & Company joint clients to surface new insights into the relationship between pre-college behaviors and first-year persistence.
The research is ongoing, but we were sufficiently excited by the results to build a special integration feature within EAB’s Student Success Collaborative platform that infuses Royall & Company data into first-year persistence analytics, and we are continuing to build out analyses and uncover insights.
Here’s a sampling of what we’ve found through this initiative:
1. Application timing
It should come as no surprise that pre-college engagement correlates to student success. More specifically, engagement during the application process correlates with how a student fares during his or her first year.
Both in schools that have rolling admission processes and in schools with precipice admissions, we found that students who file later are significantly less likely to persist to sophomore year.
2. Parent engagement
Family engagement also matters—especially for high-risk populations. We found that first-generation students whose parents were involved in the college search process had higher rates of persistence than students without parental involvement and nearly on par with the all-student average.
3. Institutional contact
Students who have no institutional contact prior to filing an application are called “stealth applicants.” We found that these students have lower retention rates than peers who connect with the school earlier in the admission process. Surprisingly, the most academically qualified stealth applicants are actually the biggest flight risks. These students are likely transferring to other institutions because they lack a deep connection to the institution.
Already applying the findings
Beyond these initial insights, some of the best news we got from our pilot schools was that frontline advisors found the insights from the pre-enrollment data to be actionable and useful.
Faculty members and professional academic advisors told us that the data helped them uncover nonacademic issues in the first few days of school, it helped them target their support toward students most in need, and it improved their ability to offer service and support to students holistically.