By Haidar M. Harmanani
Higher education institutions are collecting more data than ever before, and predictive analytics have been gaining momentum due to the abundance of tools that can make sense of data.
Predictive analytics has been defined as “the use of data, statistical analysis, and explanatory and predictive models to gain insights and act on complex issues.” They provide insight into business dynamics and facilitate optimum data use.
Analytics applications in higher education range from identifying ways to reduce the cost of education to providing early intervention to help a struggling student.
Many colleges and universities in North America have demonstrated that analytics can help advance the institution in strategic areas. For example, analytics have been proven useful in developing predictive models for enrollment management while taking into consideration the front-end admissions funnel, the prospect pool and the back end, student retention.
Learning analytics is an emerging area in higher education and is mostly concerned with the learning process. Exogenous variables are used in the model in order to assess academic progress, predict future performance, and spot potential issues. Analytics has also been successfully used in the area of fund raising and finance where it provides a better understanding for donation patterns in order to optimize alumni campaigns.
Analytics has been used as well in strategic areas such as predicting faculty research output and quality, benchmarking, computing the cost to complete a degree, facilities and resource optimizations, and administrative functions such as course offerings and optimizations.
Predictive analytics are typically combined with data visualization in order to find insights and to come up with new hypotheses and conclusions. Visualization can create unique displays that had never been seen before and facilitate the accreditation reporting process.
The analytics fever, however, has not caught up with the academic community in the Arab region. Institutions of higher education in the Arab world have shied away from the use of predictive analytics to support institutional outcomes.
While most universities in North America use analytics in some capacity, most institutional research offices in the Arab region limit the use of data to reporting and rudimentary analysis with little strategic analytics.
Modern learning management systems have embedded capabilities for learning analytics. In the UAE, according to a study by Adam Marks and Maha Al-Ali, learning analytics are mostly limited to curriculum coverage, student performance, and early alert systems.
A recent survey by Tableau shows that all but seven of the top 150-ranked national universities by U.S. News in 2016 offer formal analytics academic programs, with some offering multiple programs. For example, Harvard University offers 10 different analytics programs while Carnegie Mellon University offers 25.
In contrast, universities in the Arab region remain conservative and offer traditional computer science and business programs. To the best of our knowledge, no university in the region offers a degree in data science or analytics while very few universities offer pertinent courses.
Haidar M. Harmanani is a Professor of Computer Sciene at the Lebanese American University, Byblos, Lebanon.