Personalized and Automated: How Efficiency Benefits Students and the Institution

1. How do systems built around predictive analytics, such as the GPS at Georgia State University, support staff productivity?

The goal is really twofold. It is in the first instance obviously to help students. We can use data more effectively to do so. There are all kinds of data points that we really had at our fingertips for years that we haven’t used aggressively, proactively enough to actually benefit our students. A really simple example is that we found a grade that a student gets in the first course in what becomes his or her major is very predictive of their chances of graduating in that major. Most campuses to this day are doing nothing with that data. They’re not reaching out to the student and helping them at that point when the information is first available.

That’s the student-facing side of it, but think of the opportunities this has for staff efficiencies as well, where we often now are grappling with students and their struggles academically in our student support offices, in our advising offices, our tutoring offices. What we tend to do is progress students who got [low marks] in the first course, until they take some upper level courses and start getting D’s and F’s in those courses. By that point, the problem is much more dramatic, not only for the student but for the institution. The student may have lost scholarships or other sorts of aid and so need help from the financial aid office. The student will potentially have to switch to another major, requiring more input from our advising services and ultimately more courses to take and a longer time at the university.

The more we can help the students, the more efficient our staff can be in doing their jobs as well.

2. Building on that, how do students benefit from staff being able to focus on support and advising rather than on research?

One thing that technology can do is provide a kind of information on a large scale that is very time consuming for individual staff members to collect. One things that this predictive analytics advising system is doing at Georgia State is looking — as students register for courses each semester — for whether those courses they’ve signed up for actually apply to their program of study.

In an ideal world we’d like to think that the students will all have the wherewithal to actually pick courses that are going to advance their degree programs, but the reality is that that’s not the case. Rather than having somebody or a whole group of staff people poring over the registration records each evening, at Georgia State, we have technology that now can do that, can flag to staff if a student has signed up or if it doesn’t apply to their program of study, and that frees up time for the staff to deal with much more personal and nuanced issues in conversation with the students. The high-tech part of it has actually allowed for a grater personal touch on the part of the staff.

3. By allowing staff to focus on high-value interventions and support rather than on research and legwork, how does the institution benefit?

The greatest benefit really does start, as most things should, with the student. Our mission is first and foremost not just to enroll students and certainly not just to collect tuition dollars, but also to actually provide a valuable experience, a worthy education, and ultimately a degree to the students who pursue it and deserve a chance to get the degree.

One of the big benefits institutionally is that these systems, in combination with our staff, are increasing graduation and retention rates. That’s a huge boost to the institution.

We can really help the revenue flow of the university as well when we do our job well in supporting students. The students who persist are also the students who are going to be paying tuition and contributing to the university in the long haul. At Georgia State we enroll about 32,500-plus students every semester. If we can increase our annual retention rate — the percent of students who we hold onto from one year to the next — by a single percentage point, that’s over 300 additional students who will be in our classes, will be on our campus, will be pursuing our degrees, and students who we don’t have to spend and invest new money into recruiting.

4. Is there anything you’d like to add about the value of predictive analytics systems such as GPS and other efficiency creating projects, for students, staff and the institution?

We’ve known for a long time in higher ed that the more personal and sculpted the experience can be for the individual student, the more they will excel. Small liberal arts colleges with very small classroom sizes and lots of faculty contact with students have a very high quality experience for those students and not surprisingly the students tend to persist at higher rates and graduate at higher rates.

One of the big challenges for large-scale public education is that the financial and personal calculus has for decades been very different. We are working on a different price model, we have a different scale, there are no liberal arts colleges in the United States that have 32,000 students enrolled every semester like Georgia State, so how do you bring that same sort of personalized experience to students on a larger scale? It has long been a challenge and one of the great opportunities that we’re experiencing now, and it’ll only get better in the years ahead, is using technology, using data to be able to give students a close approximation for some of those sorts of experiences. In a small liberal arts college where you register for classes, you might have an advisor who’s looking at each and every class you choose, a mentor. Well now we’re able to do that with technology. That’s a huge step forward, not just for higher education in general but specifically for the quality of the experience that students are getting at public universities.

This interview has been edited for length.