A Model Way Of Analyzing University Performance

Graduate employability is associated with high levels of research intensity and internationalization at universities, according to a mathematical model developed by Filipino scientists.

AsianScientist (May. 10, 2021) – For universities in Southeast Asia, research output and the degree of internationalization are linked to high graduate employability, found researchers from De La Salle University (DLSU) in the Philippines. Their results were published in Cleaner Engineering and Technology.

Across industries, it’s no secret that education can significantly shape one’s employment prospects. As the lifeblood of a nation’s human resource pool, higher education institutions like universities typically aim to produce industry-ready graduates who can contribute to the country’s development.

Given how fast job markets can evolve, universities must furnish the right mix of technical knowledge, soft skills and values for graduates to thrive in their chosen careers. Clearly, a holistic experience is key, but the means to achieve this are not so clear-cut. University decision-makers often have to consider a dizzying array of strategies, ranging from modifying instructional techniques and curriculum content to organizing exchange programs and internship arrangements abroad.

Though these facets all contribute to an enriching university experience, there has been a lack of definitive evidence on how exactly these strategies improve employment prospects. To fill this data gap, a team from DLSU trained a machine learning (ML) model to analyze the institutional attributes associated with graduate employability among universities in Southeast Asia, with data pulled from the 2020 Quacquarelli Symonds Asian University Rankings.

Quickly sifting through the data, their ML model groups together different metrics—for example, the number of faculty PhD-holders, research papers per faculty and citations per paper—and validates whether these metrics accurately predict for high employer reputation.

While the model itself contains a complex series of mathematical equations, the results are expressed in a much simpler way in the form of if-then rules. These statements make up a readily interpretable classification scheme, so universities can easily be sorted as satisfying those rules or not.

According to the study, attributes that strongly correlated with high employability were number of papers per faculty, the presence of an international research network and faculty-to-student ratio. Interestingly, these results put quite a premium on research productivity and collaboration.

Though knowledge generation has only recently gained traction among younger institutions and emerging economies, the team’s findings could prompt a closer look at university priorities moving forward.

“The model can identify patterns in the data which may not be obvious or intuitive to a human. The rules identified by the model can then be used by universities to develop strategies which will help them improve institutional performance,” said corresponding author Professor Kathleen Aviso from DLSU. “We all need to ensure that our graduates will be competitive once they complete their degrees.”

While the authors cautioned that yet-hidden factors could influence these links, the model provides a data-driven starting point for universities to more effectively cultivate future leaders in different fields.

The article can be found at: Aviso et al. (2021) What university attributes predict for graduate employability?


Copyright: Asian Scientist Magazine; Photo: Shutterstock.
Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.

Asian Scientist Magazine is an award-winning science and technology magazine that highlights R&D news stories from Asia to a global audience. The magazine is published by Singapore-headquartered Wildtype Media Group.

Related Stories from Asian Scientist