Predictive analytics – once the domain of insurance companies and sports teams – is now being harnessed by universities with struggling students.

Big data is hiding behind your music recommendations on Spotify, the movies you’re shown on Netflix, and the products “you might also like” on Amazon. The technology uses predictive analytics – collecting data and analysing it to foresee future trends and consumer behaviours. Now, universities are starting to tap into the powers of predictive analytics to help their students graduate.

Over the last decade, graduation rates across Canada have plateaued. According to the Globe and Mail, up to 20 per cent of students in Canadian universities drop out and never return to any postsecondary program, while 20 to 50 per cent drop out of their initial program.

Dropping out or even just delaying graduation can be financially disastrous for students. But universities suffer too, when their graduation rates are low.

In Canada, provincial government grants are partially based on how well universities perform – meaning that if a university graduates fewer students, it’s at risk of losing grant money. As Canada’s population ages, the pool of young people applying to university is shrinking too. If universities want to stay afloat, they have to focus on boosting graduation rates.

That’s where predictive analytics companies come in.

In the last few years, big data companies have begun specializing in predictive analytics for higher education, partnering with around 200 universities mostly in the U.S.. These companies collect data from current university students, feeding their information into software that can predict the students’ likelihood of graduating on time.

One of the most notable (and noted) examples of a university successfully using big data is Georgia State University, where researchers spent four years tracking the grades and test scores of GSU students, and using predictive analytics to make forecasts about their chance of graduating,

As soon as a student showed warning signs – like taking the wrong prerequisites for their program, or doing poorly in a foundational course for their major – an adviser was alerted, who would reach out to the student and get them back on track.

Traditionally, advisers only monitor students’ GPAs, with the assumption that a grade of C or D in an intro-level course is nothing but a harmless outlier for an otherwise straight-A student. In reality, data analysts Civitas Learning have found that a student’s chance of graduating took a sharp drop when the student got less than a B in a foundational course in their major, reports the New York Times.

The outcome? According to Fortune, the number of students graduating from GSU increased by 30 per cent – and the students that graduate are spending less time and money to earn a degree. STEM degrees awarded to Black students increased by 69%, to Black male students by 111% and to Hispanic students by 226%. The average time to complete a bachelor’s degree decreased by over half a semester; meaning that the Class of 2016 saved $15 million in tuition and fees.

Grades and test scores are just the tip of the data-collection iceberg. Hypothetically, everything is fair game: personal and demographic information, the courses students choose, how often they see advisors, whether they hire tutors, how often they check online learning modules, and each time they access a journal or check out a book.

But as universities collect greater amounts of personal data, there are worries that the information could be used for racial or economic profiling, and “weeding out” weak students from a program.

Regardless of whether universities use big data responsibly or irresponsibly, it’s racialized and low-income students who are most impacted.

At its best, predictive analytics allows for specialized, immediate support for students most at risk of dropping out. Many racialized and low-income students are the first in their families to go to university or college. Without parents who have experience navigating higher ed, these students are more likely to make decisions (about which classes to take, which major to choose, and when to ignore warning signs of bad grades) that will lead them to drop out.

Rather than waiting for a student to drop out and then wondering what went wrong, predictive analytics allows for a university to help every student reach the finish line in real time. Now it’s up to universities to harness its power responsibly.  

Looking to add predictive analytics capabilities to your university or college’s tech software? Fiscot can help.

Using the power of Microsoft Dynamics AX, Fiscot offers you cutting-edge predictive analytics technology so you can start making the most of your data today.

Take a look at our Student Inquiry Management system for starters.

Image source: “Big_Data_Prob” by KamiPhuc on Flickr. Used under Creative Commons license.