Big data, predictive modelling, and international student recruitment

Earlier this year, we began to consider how better management and analysis of large volumes of data could improve student services and recruitment. In the months since, the concepts of “data mining” and “big data” have become more and more familiar in many business sectors. In this post, we will take a closer look at data analytics - and predictive modelling in particular - through a series of excerpts from a feature interview with Dr Daniel Guhr, managing director of Illuminate Consulting Group (ICG). Big data refers to information sets that are so large and complex that they cannot be properly managed with conventional database applications. But the promise of coming to grips with such large data pools is that - to the extent they can be effectively captured, stored, and analysed - they can yield important new insights that can help improve marketing, services, or operations. This process is sometimes referred to as data mining, which the research firm Gartner Inc. defines as:

“The process of discovering meaningful correlations, patterns, and trends by sifting through large amounts of data stored in repositories, and by using pattern recognition technologies, as well as statistical and mathematical techniques.”

New ideas and tools for exploring huge volumes of information have emerged to help tap into the competitive advantages that can be derived from better data analysis, and there are a variety of applications for these new methods in higher education. One of the more interesting approaches, particularly for recruiters, is predictive modelling. The consulting firm Noel-Levitz defines the term as follows:

“Predictive modelling uses statistical analysis of past behavior to simulate future results… A predictive model not only can enhance your primary market recruitment, but also identify how your institution can effectively and deliberately target students and populations in your secondary and tertiary markets - leading to increased enrolment numbers via a more efficient recruitment process. These models inform and affect not only communication strategies, but also territory management and travel… Such a tool, therefore, takes travel - notoriously a low return on investment venture - and improves this return significantly by placing statistical data and modelling at the center of territory management.”

Dr Daniel Guhr has extensive research and consulting experience, both in academia and the private sector. In the following series of excerpts from an extended interview we conducted earlier this year, he describes one such data analysis tool, PRISM, its applications in analysing student performance, and how those data-based observations can be used to improve recruitment and admissions processes. In the second part of our interview, Dr Guhr explains further how data analysis can allow institutions to build profiles of students who are most likely to succeed in their studies, and how this in turn opens the door to “recruiting for retention.” “What you have is essentially a concept of what a good student should look like,” Dr Guhr notes of the data-based student profile described above. As the interview continues, he notes that this profile can be applied to the recruitment efforts of institutions as well as agents. The end goal, he argues, is to focus recruitment - both for the educator and the agent - on not only meeting enrolment targets but on recruiting students with a better chance of success, and therefore a better prospect of retention and graduation. The interview concludes with a discussion of how predictive modelling can be used across various recruitment channels to better manage the recruitment process. In the case of agent channels, Dr Guhr reflects on how predictive models can provide a basis for incentives, both formal and informal, for agencies to target recruitment to students that fit institutional profiles for student success.