When I talk with institutions about proactively, intrusively, and holistically supporting students, a common response is something like, “Oh yes—we have an early alert system for that!”
I’m not here to bash early alert systems – indeed, I think they are a key component of any good student success strategy – but I am here to tell you that they are not proactive, intrusive, or holistic.
What do we mean by “early alert?”
The concept of early alert is a good one: identifying at-risk students facing challenges like academic struggles, attrition, or personal issues. They rely on a few things to work effectively. First, alerts need to be triggered, raising a concern or “flagging” a student. Second, someone needs to respond to that flag to reach out to the student and provide support. Third, all this usually entails a system for submitting, responding to, and tracking flagged students.
The predominant method for flagging students is through faculty alerts, which is a logical mechanism given that, in most cases, faculty have more interaction with students than any other constituency on campus (It is for this reason that any student success strategy that excludes faculty is missing a pretty massive opportunity.)
As an example, if I’m teaching an intro psychology course, and I notice a student has missed three consecutive classes, I’d submit a flag through the early alert system to let whoever is responsible know that they should reach out to that student. Typically, that outreach is conducted by someone in an advising/coaching capacity, and occasionally that role is specifically designated as an “early alert specialist” or something to that effect.
Some early alert systems are data-based. In these cases, using predictive analytics, institutions know up-front that certain factors are predictive of success. Using those models, they can identify groups of students who, without outreach and support, are less likely to be successful than their peers. Using these data as the “flag,” institutions can then effort to connect with the students who might need some help.
There are several other tools that are needed to make early alert systems work well. For example, it’s good to train faculty on how to raise flags, not just so they can operate the technical process, but so they know what types of things to look out for. Well-established communication templates can help early alert staff communicate with students in a way that is welcoming and avoids looking punitive.
In the end, early alert systems vary significantly across individual institutions. In some cases, they are the centerpiece of the student success strategy. In others, they are simply one part of a broad effort to better support students. Without going down too many rabbit holes, if you’d like to learn more, there’s a nice report, published by Hanover Research, that gives a decent lay of the land (as the land laid a decade ago, at least).
Do early alert systems work?
In a study by Kozeracki et al. (2021), it was found that early intervention systems that utilized faculty flags resulted in significant improvements in retention rates among at-risk students. Faculty were able to identify struggling students early, which allowed for the timely provision of academic support services. These students were less likely to withdraw from courses and were more likely to persist toward graduation.
A study by Ferris et al. (2020) showed that, in addition to academic issues, early alert helped address some non-academic barriers as well. Students who were flagged by faculty for reasons such as personal challenges or mental health concerns benefitted from being connected to counseling and other campus resources.
In 2006, George Kuh and several other great researchers in our field published a review called “What matters to student success.” Indeed, they do mention a broad family of “early warning systems” (which includes, but is not limited to early alert) as being critical, particularly for students who are struggling or who have multiple challenges to overcome.
However, one of my favorite things about Kuh et al.’s review is their continued finding that the effectiveness of most interventions is mixed. The results vary because it’s not about whether you have living-learning communities, a first-year experience, or any one of the myriad “best practices” we find in higher education – it’s how they are applied. I would argue that the same is true of early alert. The mere presence or absence of a system is far less important than how well it’s communicated, supported, and adhered regarding its efficacy for improving student success. This need for effective organization, implementation, and integration is just one of a few challenges that come with early alert.
Challenges to early alert systems
If you’ve ever tried to adopt any system at your institution, you know that even the best tool can fall to pieces if not implemented effectively. Integrating such approaches into the day-to-day work of improving student success is challenging in the best cases, but especially when trying to include those for whom this work is not in their wheelhouse of expertise, experience, interest, or responsibility (cough…faculty…cough). Thus, one of the main challenges to early alert platforms is the significant effort required to do it right. This includes early, effective training to introduce the platform, process, etc., but also ongoing efforts to give feedback, provide new resources, acknowledge successes and challenges, and so on.
There are some personal issues that arise if early alert – particularly flagging students – is not handled effectively. When speaking with a close friend, who also happens to be an award-winning faculty member truly committed to the success and learning of his students, he mentioned that he’s a little skittish about submitting flags, as he feels it can be intrusive into students' lives. Moreover, he said that some students have felt like he “ratted them out” when flagging them.
More broadly, when relying on faculty flags, early alert may actually exacerbate equity gaps in our student support. Research has suggested that faculty may be more likely to flag certain students based on implicit biases, such as race or gender, rather than solely on academic performance (Farkas, 2020). This could result in inequitable treatment of students and may undermine the effectiveness of the early alert system. Once again, the need to have effective training and implementation – e.g., providing objective criteria when identifying students - is critical.
The Bigger Picture
Perhaps the biggest challenge is that early alert systems represent a better version of an ineffective strategy, rather than the revolutionary changes higher education needs to improve student success. Initiatives like Guided Pathways and Holistic Student Support have shown us that if we really want to improve student success – particularly for traditionally underserved populations – we need to restructure the ways we understand and support student success.
The downfall of all extant student success initiatives is that they are reactive. Most institutions have a TON of resources, ranging from advising, to tutoring centers, to a host of other cocurricular programs. The problem is that that they wait for students to engage. They are reactive, relying on the student to understand and appreciate their value before using them to improve their success.
Similarly, early alert initiatives rely on faculty to acknowledge a student’s challenge and submit a flag. Indeed, effective training can enhance the quality and frequency of faculty efforts, but they are still reacting to a challenge. By the time a student has missed three classes, there’s a good chance that the process of attrition has already begun.
Moreover, early alert still relies on student engagement. Faculty may flag students who are struggling academically, but if those students do not engage with the intervention process, the effectiveness of the system is diminished. A study by Bain et al. (2018) showed that while students who received intervention support showed improved outcomes, students who did not respond to outreach efforts did not experience the same level of success.
Conclusion
Ultimately, I come back to the point that early alert can be a powerful tool for student success, but it’s only that – a tool, a component, a part of what we should be doing to better support students. Certainly, we need the capacity to respond to issues that arise, and faculty can and should play a central part in that. Again, given that they are likely the closest observers of students and have the best chance of seeing a critical issue, their the best resource we have. However, that resource still needs to be convinced, trained, and supported to be effective.
But if we continue to rely on waiting until problems arise – whether those are disclosed by students or flagged by faculty – we’re relying on lagging an ineffective mechanism. Thus, if you’re thinking about how to best integrate early alert, my final recommendations would be this:
1). What is your plan for organizing, training, and integrating early alert into a regular part of what faculty do and students expect?
2). In addition to early alert, how are you proactively supporting students?
There’s an old adage in student affairs: supporting students before they even know they need it. Initiatives like Guided Pathways and SSIPP have started to help schools think in these ways, but early alert is a remnant of responsive and reactive approaches.
References
Bain, S. D., Brown, S., & Smith, E. F. (2018). Student engagement and retention: The role of early alert systems. Journal of College Student Retention: Research, Theory & Practice, 19(4), 493-515.
Farkas, G. (2020). Bias and inequality in the early alert system: A case study. Educational Policy, 34(5), 745-763.
Ferris, E. B., Lewis, R. R., & Thompson, R. H. (2020). Integrating mental health into early alert systems. Journal of College Student Development, 61(6), 712-728.
Kozeracki, C. A., Huerta, M., & Lantz, A. (2021). The impact of early alert systems on student retention. Higher Education Journal, 74(2), 159-177.