With overall college enrollment on the decline in the U.S. (NSC, 2014), competition among colleges for top students is aggressive. As technology has streamlined and simplified the college application process, more students are choosing to apply to multiple schools, with over three-quarters of students applying to at least three schools (NACAC, 2013). As a result, traditional metrics for creating acceptance targets and calculating student yield are becoming less reliable, and it is no longer sufficient just to bring in large numbers of applications. The pressure is on to ensure that as many acceptances as possible materialize into students filling seats in classrooms. At the same time, budgets are tight. State cutbacks have hit public institutions hard, and many private schools saw sharp declines in their endowments and donations during the recession.
Admissions officers are thus looking seriously at how they can best and most cost-effectively appeal to the current generation of largely technically savvy, pro-social college applicants. One popular avenue is the use of social media, which is attractive both in its low cost and broad reach. However, little concrete data exists on the efficacy of social media marketing for the purpose of college recruitment.
In 2011, the University of California, Santa Barbara (UCSB) began inviting prospective students to connect through a closed Facebook community as a part of our overall recruitment strategy. In this article we review how we assessed the effectiveness of that tool, specifically on the transfer student population, what we found in our assessment, and what we’ve learned over four years of using the tool.
Why We Suspect Social Media Works
Though reports of its effectiveness are largely anecdotal, it is little wonder that the possibilities of social media capture the imaginations of college recruiters. Social media allows for a much more meaningful and interactive recruitment experience than many of the more traditional channels. Rather than simply filling students’ mailboxes with pages of colorful and expensive marketing materials that they may never look at, and hoping for the best, social media offers the opportunity to invite prospective students into a dialog – get them thinking about and asking questions about the school. In addition, whether or not they engage and their level of engagement can be monitored.
Social media also offers a great deal of flexibility. There are countless ways to use it in recruitment – targeted ads, webinars, targeted communities, activities such as web scavenger hunts, surveys, hashtags, etc. And in many of these cases the students themselves become part of the promotional process, a source of information that other students might perceive as less biased and more reliable than the school itself. A 2009 study of college students in Texas found students’ use of social media to be positively associated with attitudes of social trust (Valenzuela, Park, & Kee), so it makes sense that recruiters would wish to tap into this seemingly trustworthy resource to generate interest in their schools.
Moreover, for decades researchers have observed the relationship between student engagement and integration, and college persistence (Cabrera, Castaneda, Nora, & Hengstler, 1992). Social media offers the first meaningful opportunity for colleges to extend that to prospective students, encouraging them to create relationships early, promote a sense of belonging, and develop loyalty to the school and their peers who attend or plan to attend.
Why Look at Transfer Students?
These possibilities are perhaps even more exciting when considered in terms of transfer students. Transfer students have long been shown to have a particularly difficult time adjusting to life on a new campus as they suffer from “transfer shock” from which they may have difficulty recovering, impacting not only the outcomes tracked and reported by the university, such as GPA, retention, and graduation, but also less easily measured outcomes such as level of involvement and social satisfaction (Laanan, 2007). By engaging transfer applicants prior to the point of matriculation, schools may be able not only to encourage their decision to attend, thus improving yield rates, but also ease their initial transition and ultimately improve their long-term outcomes.
Background of UCSB’s Assessment
The purpose of UCSB’s assessment was to examine the effectiveness of our closed Facebook community in the recruitment of transfer students. Specifically, we wanted to know: 1) Is the UCSB Facebook community effective as a recruitment tool? 2) If so, how effective? 3) Is it more effective for certain subpopulations than for others?
The study examined 4415 prospective junior-level transfer students admitted for Fall, 2012. Approximately half these students were randomly selected to receive an invitation to the university’s official Facebook community (invitation group), and half were not invited (control group). Descriptive statistics for the sample can be seen in Table 1.
The community functioned as a Facebook “app,” and was available only by invitation (not discoverable by search). Those who were invited to join the community received their initial invitation via email several days after receiving their offer of admission. It was promoted as a convenient and secure place to connect with other prospective and current students and learn more about the university.
Within the community, interaction was driven completely by students, who could form sub-communities, organize meetups, post photos, etc. Monitoring by university administrators was minimal, and only for the purpose of maintaining a safe and open environment. No “seed” posts were planted by staff.
Statistical Analyses and Results
Analyses were conducted using SPSS version 22.0. All data were analyzed within a treatment-on-the-treated (TOT) framework, which accounts for treatment uptake. In other words, it statistically controls for the fact that only those who actually joined the Facebook community, in this case 27.6% of those who were invited, could actually benefit from it. The Wald estimator was used to calculate a local average treatment effect (LATE), and two-stage least squares (2SLS) regression was used to conduct a more robust analysis.
The Wald estimator revealed an overall effect size of 5.3%, indicating that transfer students who participated in the Facebook community matriculated at a rate 5.3% higher than students who were not invited to participate, but otherwise likely would have.
Three different 2SLS models were tested. Each of these models incorporated four covariates – gender, ethnicity, highest parent education level, and residency classification – along with the students’ participation in the Facebook community. The model with the best overall fit incorporated interaction effects between the covariates and whether the students were invited to join the community, F(10, 4113) = 62.34, p < .001, R2 = .132. This indicates that different types of students responded differently to the invitation. Specifically, students whose parents had some college education were more likely to accept the invitation, and students who lived further away from the campus were less likely to accept the invitation. However, among students who participated in the community, benefits seemed to be roughly equal for all types of students.
The full results of the analysis can be seen in Table 2. After controlling for all other variables in the model, participation category (i.e. whether or not the student participated in the Facebook community) showed a significant impact on students’ matriculation patterns, β = .35, t(4104) = 15.44, p < .001. Not surprisingly, other variables that were found to significantly influence matriculation included ethnicity, parent education level, and residency classification.
What We Learned from the Assessment
The low overall participation in the social media community even among those who were invited (27.6%), as well as the modest model R2, indicate that most students make their matriculation decisions based on criteria that were beyond the scope of this assessment, and that social media networking opportunities have a limited ability to influence students’ behavior. In terms of college matriculation patterns, this makes sense. Any enrollment services veteran knows that most students applying to multiple colleges will have preferences among those to which they applied, and will accept admission to the school they most prefer among those to which they were offered admission. Thus, the students most likely to be influenced in their choice of college by participation in a social media community are those who have not yet made a definitive choice based on other criteria. However, the results do indicate that for students in this situation, this tool can have a statistically significant impact. When one considers a 5.3% difference in yield across a large pool of admits, this can mean a substantial difference in the overall number of matriculates.
The results may also have been stronger had we looked at freshmen rather than transfer applicants. Though transfers potentially stand to benefit more from social media in terms of a smooth transition to campus, they may also be more likely to have made up their minds about where they wish to go before ever having the opportunity to participate.
A Few Other Things We’ve Learned
In addition to what we learned from our assessment, we have also picked up quite a bit from four years of observing student interactions within the community. Here are some highlights:
- The transition to a new college follows a predictable cycle. The pattern of discussion within the community is similar from year to year. Initially students are excited about their acceptance and meeting new friends. They soon move on to more pragmatic concerns, like financial aid, housing, and getting classes. As orientation approaches, they begin to organize meet-ups and share tips for navigating the university environment. Near the beginning of school, we see a mix of practical and social concerns. The students tend to be great about supporting each other, regardless of what stage they are at. After a year or two, once you know what issues to expect at what points, you can also have staff in key areas on deck to assist them.
- A social media community is a recruitment tool, not a retention tool. Though students are really excited to connect in this way before they come to campus, participation in the community drops off sharply once they arrive. Though they continue to use it occasionally, particularly during the first year, for things like selling textbooks and finding carpools home, even this incidental use eventually tapers off.
- Prospective students want to hear from current students. Even though use among current students is low, you don’t want to purge them from the community. During our first year using this tool, we included only prospective and no current students, which generated a lot of negative feedback. Prospective students really want to hear what the campus is like from those who have already experienced it, and there are always a few current students ready to jump in and answer their questions.
- Students don’t group themselves in the way school staff might expect. When we initially launched the community in 2011, we seeded it with a number of interest-based sub-communities focused around topics like majors, residence halls, and student organizations. However, all of the most popular sub-communities were since created by students, and cover everything from “Laughing” to “Virginia Wolfe.” To date students have created well over 3000 sub-communities.
- Tone matters. If you have staff putting out any kind of messaging within the community, it shouldn’t be cutesy, patronizing, or institutional. Students want to be talked to straight and like adults. Cheekiness is encouraged!
- They don’t want to win an iPad. If you use the community to solicit feedback (e.g. “Fill out this survey for a chance to win…”), don’t expect the opportunity for big prizes to motivate students. Better to appeal to their sense of community and desire to contribute. And keep those surveys short!
- Students love connecting with each other online! Regardless of the ultimate impact on recruitment, it’s clear that prospective students enjoy “meeting” one another before coming to campus, commiserating with peers who are struggling with some of the same doubts and questions, getting the inside scoop on campus life, and feeling like they are part of the larger campus community.
So the ultimate conclusion we were able to draw from our study was that, yes, social media networking communities can have a positive impact on recruitment, particularly for students who are on the fence. However, it is important to understand that students are going to connect this way with or without us, creating their own social media communities if we don’t provide them. The effectiveness of social media as a recruitment tool may soon be a moot point, as having a strong social media presence will be necessary for institutions simply to keep pace with their peers and student expectations.
- Angrist, J. D., & Imbens, G. W. (1995). Two-stage least squares estimation of average causal effects in models with variable treatment intensity. Journal of the American Statistical Association, 90(430), 431-442.
- Bowden, R. J., & Turkington, D. A. (1990). Instrumental Variables. Cambridge, UK: Cambridge University Press.
- Cabrera, A. F., Castaneda, M. B., Nora, A., & Hengstler, D. (1992). The convergence between two theories of college persistence. The Journal of Higher Education, 63(2), 143-164.
- Hoover, E. (2010, November 5). Application inflation. The Chronicle of Higher Education.
- Laanan, F. S. (2001). Transfer student adjustment. New Directions for Community Colleges, 2001, 5–13.
- Laanan, F. S. (2007). Studying transfer students – part II: Dimensions of transfer students’ adjustment. Community College Journal of Research and Practice, 31(1), 37-59.
- Murnane, R. J., & Willett, J. B. (2011). Methods matter: Improving causal inference in educational and social science research. New York, NY: Oxford University Press.
- National Association for College Admission Counseling. (2013). 2013 state of college admission. Arlington, VA: Clinedinst, M. E., Hurley, S. F., & Hawkins, D. A.
- National Student Clearinghouse. (2014). Current term enrollment estimates – Spring 2014. Herndon, VA.
- Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson Education.
- Valenzuela, S., Park, N. & Kee, K. F. (2009), Is there social capital in a social network site?: Facebook use and college students’ life satisfaction, trust, and participation. Journal of Computer-Mediated Communication, 14: 875–901.
Leesa M. Beck is the University Registrar at the University of California, Santa Barbara Prior to serving in her current role as University Registrar at the University of California, Santa Barbara, Leesa Beck spent a decade working in enrollment services technology. She has played an integral role in SIS implementations at two separate institutions, and served as project manager or functional lead on numerous successful systems projects. She is currently writing her dissertation on the use of social media in enrollment services, and expects to receive her PhD from UCSB later this year. Leesa received her BA from Cornell University and her MBA from Pepperdine University.
Author’s Note: I would like to acknowledge Keri Bradford, Student Affairs Social Media Coordinator, who co-authored and co-presented the original presentation at the PACRAO Annual Meeting in Portland, Oregon.