Knowing when to target students with timely academic learning support: Not a minefield with data mining
Concise paper
Elizabeth McCarthy
University of Southern Queensland
@elzbthmccrthy
Catch this session
Monday 4 December, 1.50pm – 2.10pm
Stream 4
Room L209
Abstract
The strategic scheduling of timely engagement opportunities with academic learning support, targeting specific student cohorts requires intentional, informed and coordinated planning. Currently these timing decisions appear to be made with a limited student focus, which considers individual course units only as opposed to having an awareness of the schedule constraints imposed by the students’ full course workload. Hence, in order to respect the full student academic workload, and maximise the quantity and quality of opportunities for students to engage with learning advisors, a means to capture and work with the composition and distribution of student full workload is needed. A data mining approach is proposed in this concise paper, where public domain information accessed from the back end HTML language of course unit information webpages is collected and consolidated in graphical form. The resulting visualisation of the students’ academic learning activities provides a quick and convenient means for academics to make informed scheduling decisions. The case study presented describes the implementation of the data mining in the context of discipline specific academic learning advisors at the University of Southern Queensland servicing three campuses under the ‘One-University’ model.
About the Author
Elizabeth McCarthy
Elizabeth McCarthy is a learning advisor, specialising in mathematics skills, and an academic in the mathematics and engineering disciplines with experience of 10 years. She is a mechatronics engineering, machine learning and mathematics enthusiast who is currently working towards her PhD project. For fun, she enjoys coding data science apps and tools to improve access to data for decision making purposes.