To the woe of underachievers everywhere, gone are the days when clever college students could hide a bad grade by getting to the mailbox first.
Today’s institutions of higher education have a wide array of predictive analytics tools at their disposal, which means there are more eyes than ever on those C’s and D’s. Research shows that examining student data helps institutions better identify and support those who are struggling. This in turn improves not only learners’ outcomes, but also reduces the increasingly high cost of attrition being faced by today’s universities.
BREAKING DOWN BIG DATA
First of all, let’s break down what university leaders and data experts really mean when they say “big data.” An article in Trusteeship Magazine draws a distinction between the two types of predictive analytics most often at play in higher education settings: institutional analytics and learning analytics.
The goal of institutional analytics, says Phillip D. Long, deputy director of the Center for Teaching and Learning/Continuing and Innovative Education at the University of Texas at Austin, is to “move the needle on some higher-order institutional metrics like attrition rates or year-to-year completion rates.”
In the world of higher education, institutional analytics are key, especially for universities struggling with funding streams. A 2012 report on the effect of student attrition nationwide revealed some startling truths about the impact of unfinished degrees on school budgets, especially when it comes to public institutions: “Attrition accounted for 33 percent of all estimated expenditures at public two-year institutions, 13 percent at public four-year institutions, and 9 percent at private four-year institutions,” the study reported.
This type of big-picture data analysis is vital for higher education leaders seeking to understand the scope of the attrition issue. However, UT Austin’s Long also emphasizes the power of learning analytics, or data that helps decrease student drop-out rates by providing instructors and learners with information about their progress. Says Long, “learning analytics is focused at the level of the individual learner and on giving learners actionable information to make their decisions about study within a given course or set of courses.”
Taken together, these two analytical approaches constitute “big data,” in this case a growing body of information on academic performance and student engagement that gives universities the insight they need to improve flagging student graduation rates, and ultimately save their bottom line.
THE COLLEGE COMPLETION CRISIS
While some big data critics point to its inherent privacy concerns, the scope of the problem this kind of info collection is seeking to correct is humbling. A recent New York Times article reported that “a little less than half of the nation’s students graduate in four years; given two more years to get the job done, the percentage rises to only about 60 percent.”
These staggeringly low graduation rates mean that almost half of our nation’s college students are leaving school with not only a big chunk of debt, but also no degree to show for it. The impact is indisputable: whether you’re talking about the quality of life for non-degreed young people seeking meaningful, well- paying work, or the financial health of higher education institutions dealing with a dropout crisis, falling retention is not an issue that should be swept under the rug.
University leaders have begun to use big data to understand not only academic performance, but also other metrics – like engagement and interactions – that may predict whether or not a student will drop out. Online courses allow instructors and school administrators to capture and analyze a great deal of information through online Learning Management Systems (LMS), according to a paper published in 2011 in EDUCAUSE Review. However, the study’s authors noted that “most LMS analytics models do not capture activity by online learners outside of an LMS (i.e., in Facebook, Twitter, or blogs).”
This has begun to shift. A 2015 survey by Kaplan found that 40 percent of admissions officers were using social media as a predictive analytics tool for determining acceptances. Likewise, big data researchers are now testing methods of gathering info on current students to get an accurate picture of who may be most vulnerable to dropping out.
Dr. Sudha Ram of the University of Arizona has focused her analysis on the social interactions of 30,000 college freshman over three years, according to the New York Times article. Ram collected and studied student data, including how many times they swiped their ID card to eat meals, purchase supplies, and access recreational facilities. While this approach may feel a bit like Big Brother U, Dr. Ram notes that retention goes beyond academic performance. Put simply, states Ram, “if they are not socially integrated, they drop out.”
[RELATED]: CARNEGIE MELLON—AN ONLINE SCHEDULING SOLUTION FOR ACADEMIC ADVISORS
PRACTICAL APPLICATIONS OF BIG DATA FOR BETTER OUTCOMES
Think the big data revolution in the field of higher education is just a pipe dream? Then it’s time to acknowledge some documented outcomes that prove this type of analysis is a game-changer. A paper published in the Journal of Asynchronous Learning Networks by two authors from the Lone Star College System opened with this startling prediction: “It is estimated that a 1 percent improvement in graduation and completion rates for higher education institutions in the Houston region would boost personal income in the area by $4.2 billion annually.”
The paper went on to detail the application of a comprehensive range of data tracking tools used at Lone Star, a community college network in Texas with 90,000 enrolled students. By capturing student data, and then acting on the findings through tools like online advising and online appointment scheduling, student engagement exploded.
Due to increased demand, “[the program] expanded its service hours by 53 percent – increasing its availability in the evenings and offering services seven days a week. In fact, online advising hours averaged 15 hours more per week than what face-to-face students had access to through their campus advisers.”
And while big data may be more easily integrated into online learning environments, many schools with face-to-face courses, like Virginia Commonwealth University, have realized benefits from using predictive analytics. By identifying students with academic “red flags” like missed classes or low grades on essential foundational courses, administrators can enact early interventions that keep students in school with the support they need. In fact, VCU was able to improve course completion rates by 16 percent after only a single semester of using big data to identify students in need of interventions, according to a report by The Washington Post.
Now that’s some news to write home about.