By Bill DeBaun, Director of Data and Evaluation
Last year at SXSWedu in Austin, TX, it was encouraging to hear representatives from a number of higher education institutions talk about their institutional belief that their commitment to students cannot end when the acceptance letters are sent out. Across multiple panels, these representatives reinforced the idea of getting students from matriculation to graduation being a shared responsibility of, and benefit for, students and institutions. Even more gratifying, as someone who extols the virtues of data and evaluation in the college access and success field, was hearing that universities were committing more resources toward being data-driven for student success.
The Association of Public and Land-grant Universities (APLU) and the Institute for Higher Education Policy (IHEP) collaborated on a series of 14 case studies of institutions that are “using data to increase student success.” These institutions, which vary in geography, size, and progression toward their goals, have all made a commitment to employing student-level data to better serve students and meet and surpass institutional goals.
The case studies (none of which is longer than three pages for the time-crunched reader) each follow the same outline: institutional context, using data to improve student outcomes, results, and lessons learned. They make for insightful reads that show what forward-thinking institutions are doing to fulfill their institutional missions and do better by students.
Across the 14 case studies are some common lessons applicable to NCAN members’ work: avoid putting data in a silo and instead make it more available to more users so they can put it to work; be thoughtful about institutional goals and priorities and the available levers for achieving them; and engage multiple stakeholders, including students, in the process of developing a data system. Additionally, many of the institutions profiled in the case studies are employing predictive analytics that examine historical student data, identify decision points and barriers that derail their retention or success, track each student’s progress at these points, and send an early warning to advisors and other staff to intervene and provide guidance before a student gets off-track.
The 14 case studies cover these institutions:
NCAN members who are near any of these institutions would be well-served by reading that particular case study and seeing if there is any room for collaboration. Higher education universities can be, and often are, important and powerful partners in fulfilling our field’s mission. Below, we provide some of the case study highlights.
Colorado State University, whose student body is 20 percent low-income students or students of color and 25 percent first-generation, believes that “data should not be ‘owned’ by any one office on campus. Instead, data should be institutionally ‘owned’ and accessible, accurate, and timely while individual offices serve as stewards to facilitate appropriate data use, interpretation, and understanding. Data, turned into actionable information, have propelled institutional change at CSU.” Officials there have been using the predictive analytics mentioned above to identify at-risk students and flag them for advisor interventions. This system engaged faculty by asking them to identify students who might have trouble in a given course based on their early performance. CSU’s data “indicate a high level of accuracy in identifying students at risk of academic difficulty” through this process. Overall, CSU has increased completion rates in early coursework in the STEM fields and, more broadly, increased overall graduation rates, second-year retention, and first-year credit completion.
Georgia State University is often held up as a model for data-driven improvements in postsecondary student success. This is for good reason. The case study notes that 10 years ago, GSU’s overall graduation rate was less than 33 percent, and graduation rates for students of color were even lower (22% for Latino students and 29% for African-American students). There were also substantial completion gaps between students of different incomes. GSU worked to remedy this by asking, “What if [we] were to take the troubling student outcomes data seriously? What if [we] were to use data to diagnose the reasons that students were dropping out and to design and implement innovative interventions to help keep students on track to complete their degrees?” They worked with the Education Advisory Board and 10 years' worth of student-level data to build an advising platform with over 800 obstacles to student success that tracks student-level progress in real-time. The case study notes that since implementing this system in 2013, “the class of 2015 took, on average, about half a semester less to graduate than the class of 2013 — saving students almost $10 million in tuition and fees.”
Additionally, Georgia State implemented a Summer Success Academy and Panther Retention Grants that, respectively, have improved first-year student performance and consequent second-year retention and helped students to graduate or remain enrolled despite having a financial aid gap.
The results have been impressive. GSU’s overall completion rate is up 22 percentage points since 2003, and the university has completely closed completion gaps based on race/ethnicity, income, and first-generation status.
The case studies do not focus entirely on four-year universities. Miami Dade College, the largest community college in the country, gets the spotlight with a case study focused on how it has been building a data infrastructure that will allow officials to concentrate on student success.
Morgan State University, which set a 50-percent graduation rate as its goal, used data coordinated from its STAR (Student, Technology, and Retention) system, implemented by Starfish Early Alert and Starfish Connect, to improve retention, which is up for every cohort since 2010.The case study describes new efficiencies in identifying and contacting students and recording the subsequent data. That data now goes into a centralized system where a wide variety of staff, including “students, faculty, and other campus constituents such as academic deans, counselors, and athletic coaches” can make use of it.
Overall, these case studies are a useful resource for the college access and success field. Colleges and universities interested in how they can employ data to improve student success now have a set of 14 resources from which to draw information. Similarly, NCAN members and other college access and success programs can show these case studies to their postsecondary partners in an effort to spur new approaches to serve students better.