The Patterns Behind Summer Melt, and What Colleges Are Doing to Stop It

A student project at the University of Pennsylvania got Sarah Simon passionate about summer melt, the admissions term for when students who have made an enrollment deposit at a college don’t matriculate there — or, sometimes, anywhere. Ms. Simon, a political-science major and education-policy minor, noticed that most research on summer melt focused on students and nonprofits. She saw that she could make a contribution by investigating patterns of melt and how colleges combat it. The Chronicle spoke with Ms. Simon, a rising senior, about her research. The following conversation has been edited and condensed.

Q. How did you get interested in this topic?

A. I first became interested in summer melt when I participated in the Penn Public Policy Challenge, a competition that Penn puts on every year. Students are put on teams and tasked with identifying a policy problem facing Philadelphia. My team was all passionate about education, and through talking to people discovered the problem of summer melt, which is really prevalent in Philadelphia. So I came at it initially from more of a practitioner/policy maker’s point of view, but then eventually became interested in it academically as well.

Q. So did you write this paper as part of that challenge, or was that something you did later?

A. The challenge was my freshman year, and we founded a program that is aimed at ameliorating summer melt. I currently run that program, which works with recently-graduated seniors in Philadelphia and helps them matriculate.

But I also for my minor have to do a capstone project, and summer melt seemed to be the natural thing for me to research since I’m so passionate about it.

Q. How did you conduct your research?

A. Most of the research is focused on why students might melt away, or maybe how nonprofits are attempting to mitigate summer melt, but very little of the research focused on how colleges perceive the phenomenon and what solutions they’re trying to implement. So I felt like that would be a good place for me to place myself in the research. I made contact with the National Association for College Admission Counseling, and spoke to them about what I was hoping to do, and they were really willing to help me out, because it’s pertinent to their mission. I surveyed their postsecondary members, all of whom are college admissions officers, enrollment managers, about my research questions: How do they see summer melt? How does it vary by institution? And what sorts of solutions are they trying to implement? And then I did follow-up interviews with 20 admissions officers to get a more in-depth take on — Why did you implement that type of solution? How’s it going for you? — things that were harder to capture in the survey.

Q What did you find?

A. The first research question I had was, How does summer melt vary by institution, or, What are the patterns of melt? Colleges are understandably much more attuned to whether or not a student melts from their particular institution as opposed to not going to any postsecondary institution at all. And to that end institutions described sort of a dichotomy of how their melt occurs. The first is what they termed “upline melt,” and then the second was “downline melt.” Upline melt usually happened with the waitlist, when students go to a more prestigious or more selective institution than the institution at which they originally committed. The second phenomenon was downline melt, which is when students melt from their institution and go to a less prestigious, or less selective institution. That happens maybe due to finances, due to family concerns or pressures, also due to distance from home. And then the summer melt that I originally started looking at, which is when students don’t go to college at all, that also classified as downline melt.

The second research question is, What sorts of solutions are universities and colleges implementing to mitigate summer melt? And this also emerged in sort of a dichotomy, between front-end solutions, which are solutions that are done before students even deposited, or before they even show signs of melt, and back-end solutions, which are more retroactive or reactive to when students start to show warning signs that they might be melters. So front-end solutions were increasing the commitment to the university, whether that be through increasing the amount of the tuition deposit or establishing a brand identity, as well as mass communications like social media, Facebook groups for incoming freshmen, and mailing lists.

On the other side of things were the back-end solutions, and those usually happened when a red flag was sent up, signaling to admissions officers that, OK, this student might be susceptible to melt. Then usually it was admissions officers who would personally contact that vulnerable student and either identify that the student was melting or give them support so that they wouldn’t melt.

Q. What do you see as the implications for people working in admissions?

A. Institutional characteristics were really big determinants, at least anecdotally, when I spoke to admissions officers, of what sorts of solutions are going to be relevant or doable. For example large institutions — meaning land-grant publics — might have a much more difficult time with back-end solutions, mainly because their staff might not scale up to allow admissions officers to be able to personally call people who don’t register for classes. Pretty much all institutions are able to rely upon front-end solutions. They’re much more mass-outreach oriented, and also a lot of the work for the front-end solutions, like establishing a brand identity, is done before students even commit to the university. So I think everyone should pay attention to front-end solutions, but the back-end solutions are important as well, because the admissions officers I spoke to seemed to think it was the more effective way of mitigating summer melt. You’re literally stopping melt from occurring if you’re talking to students who might have just ended up not coming because no one reached out to them.

Q. I know you’re hoping to get the paper published. Where does that stand?

A. I currently have the article under review at a journal. Also, the National Association for College Admission Counseling, the group who got me the initial contacts for admissions officers, I’m going to be publishing a pretty substantially different version of the research that’s much more best-practices oriented, and that should hopefully be coming out in their fall edition.

Q. What implications do you see for future research?

A. That conceptual model that I proposed, upline versus downline melt, definitely requires more empirical or quantitative confirmation that those patterns exist. The natural question you would have after looking at my paper is: Where are the students who don’t go anywhere melting from? My research would indicate that they’re melting from community colleges, but I would be interested to see a really rigorous — probably using National Student Clearinghouse data — analysis of: What are the changes and shifts over the summer? Are students melting up? Are students melting down? And, What are the flows? Where are they coming from and going to?

Q. You’re clearly very interested in education. Are you planning to work in the field after you graduate?

A. Definitely. The odds are pretty likely that I’ll be teaching in the Indianapolis Public Schools when I graduate. I hope eventually to enter education policy making, I’d love to maybe work as a superintendent, work on a state level. I don’t think academia is the place for me, I’m much more practice-oriented, or taking those best practices from research and putting them into effect. So definitely I would love to make a career in education policy.

Beckie Supiano writes about college affordability, the job market for new graduates, and professional schools, among other things. Follow her on Twitter @becksup, or drop her a line at beckie.supiano@chronicle.com.