Using big data and predictive analytics to recruit international students

A more tech-savvy approach to student recruitment could save institutions more than just time, says Marguerite Dennis

十一月 21, 2016

In 2011, Clayton Christensen wrote The Innovative University, detailing the reasons why big disruptions would soon be coming to higher education both in the US and around the world. Globalisation, changing student enrolment profiles, emerging student markets and outdated business models were among the factors driving the need for change.

Three years ago, massive open online courses (Moocs) and flipped classrooms were heralded as the disruption that would change forever the way that knowledge is delivered. The impact of Moocs on higher education delivery will be debated for years to come. But today we know that they have not replaced students in bricks-and-mortar classrooms.

In 2015, competency-based degree programmes received a great deal of attention as the next big disruption in higher education. While gaining in popularity, competency-based programmes have not replaced the traditional way of awarding credits.

This year, several articles and conference presentations have been made on big data and predictive analytics and the impact on several aspects of higher education, from predicting which students will enrol, which students are likely to graduate, and which students will become active alumni.

With regard to student enrolment, big data and predictive analytics has the ability to construct analytical databases that can provide college administrators with speedy, actionable information in order to make smart enrolment decisions and allocate both staff time and financial resources to increase enrolments from existing markets while building new ones.

From a student’s perspective, big data and predictive analytics can be used to create a shortlist of colleges that best fit the applicant’s profile. For example, Cialfo, a Singapore-based college prep course, combines traditional tutoring with data analytics to help college applicants. Its algorithms, based on thousands of data points, can guide the applicant to those colleges and universities that best match the applicant’s profile.

With regard to predicting which students are likely to graduate, big data and predictive analytics have the ability to help college deans create a set of assumptions as to which students are likely to persist and which are likely to drop out and why.

Since most colleges and universities worldwide have created strategic plans that include increasing international student enrolments, it is reasonable to extend the benefits of big data and predictive analytics to guide strategic international planners to capture the information needed to determine why prospects from a particular country decide to apply, and enrol, and why prospects from other countries do not.

What was the application trigger point, and what behavioural information can be applied to better recruit and enrol international students? Instead of building strategic international plans based on what internal marketers have determined are the reasons why international students should apply, big data and predictive analytics can inform recruiters as to the real reasons prospective students applied and at what point in the application process they reached their decisions.

This information can inform international strategic planners and recruiters about the parts of their branding proposition that are resonating with international applicants – and which parts are not.

With this information, international strategic planners can create evidence-based international strategies and change recruitment plans in real time as opposed to waiting for end-of-the-year results. Big data information and predictive analytics can assist international recruiters to prioritise geographical markets for international recruitment and target resources accordingly. This kind of information can save colleges and universities resources allocated to agents, international conferences and fairs, and international travel.

I am not suggesting eliminating any of these parts of strategic international plans. But I am suggesting using big data and predictive analytics information to make data-driven decisions on which agents, which conferences and which countries are best to invest recruitment funds. I am suggesting making strategic international recruitment plans based on evidence.

Where to begin? I have four recommendations.

Culture shift

Taking a big data and predictive analytics approach to enrolling international students will require a shift in culture and administrative mindsets. Pushbacks are inevitable. Most often, a university’s culture will eat innovation for breakfast. That is why the support of the president, vice-chancellor, provost, dean and chief financial officer is essential.

New administrative team

The director of technology and the director of research and analytics should be invited to all international strategic planning sessions along with international recruiters and counsellors. Their assistance and expertise is necessary in order to move forward with big data and predictive analytics strategies.

Internal surveys

International marketers can supplement big data and predictive analytics information by conducting their own surveys of the international students who applied and enrolled and those who did not. What triggered and influenced their behaviour? Knowing why applicants did not enrol is as important as knowing why they did. Survey results should be compared with big data and predictive analytics information to create a database of both groups, by country, and create a profile based on this data, of applicants most likely to enrol.

Smart use of social media

Big data and predictive analytics information can, and should, be supported by social media. Prospective college students are increasingly using social media to guide and influence their college choice.

In the recently published Social Admissions Report, 76 per cent of the class of 2017 in the US have used social media as a resource when deciding where to enrol. Clearly applicants prefer to connect online. The prospectus or campus visit may once have been the bellwethers of applicants’ interests. But that is no longer the case. Rather, applicant interactions on social media sites, leaving many data trails, can provide insights into applicants’ behaviour.

Case in point. Several years ago, when I served as vice-president for enrolment and international programmes at Suffolk University in Boston, on the Friday before the Monday deposit deadline, deposits for entering students were considerably lower than what was needed to meet the “bottom line”. I discussed this with the director of research and over the weekend she phoned me and told me to relax.

Based on her investigation of several Facebook postings, the requisite number of deposits would be made by Monday. She was correct.

Last year, the University of Denver had a similar result with their private social network, Schools App. Admitted students who joined the online community before 1 May were 3.6 times more likely to deposit than those who did not join.

Facebook, Twitter, Snapchat, Instagram and YouTube are the biggest social platforms today. Information from these sites, along with historical applicant data, can provide insights for international strategic planners to target those applicants most likely to enrol and graduate. An admissions dean can follow which applicants are posting pictures and photos of the campus to Instagram, or which applicants are tweeting about their college visits.

Google searches of applicants may reveal disqualifying information. Some high school students are being coached by guidance counsellors on how to “clean up” their online presence.

The best use of information from social media sites will require input from a tech-savvy employee who “lives” on several sites and knows how to investigate and interpret information that can then be used by international strategic planners.

It is best to start small. You will need to tailor your approach to the organisational needs of your school. You will need to develop agreed-upon standards. You will need to make the case for making decisions based on evidence.

The world of higher education is transforming at a pace that invites disruptions. Is big data and predictive analytics a disruption? Is utilising actionable information to enhance international strategic planning and recruitment really a disruption? My guess is that in a short while big data and predictive analytics, along with Moocs and competency-based education, will all be considered acceptable parts of the higher education landscape.

Let’s hope the atmosphere for innovation always trumps stagnation.

Marguerite Dennis is a higher education consultant.