ARTICLE # 6 | 101
ARTICLE | # 6
Title Data Mining for Learning Analytics; does lack of engagement always mean what we think it does?
Author( s)
Ed de Quincey Theocharis Kyriacou Mark Turner
Contact e. d. quincey @ keele. ac. uk
School
School of Computing and Mathematics
Faculty
Faculty of Natural Sciences
Abstract
Context and Objectives: Learning Analytics( LA) has the potential to utilise student data to further the advancement of a personalized, supportive system of HE( Johnson et al., 2013). A number of LA systems are now being developed but there have been few studies that have analysed the usage of Virtual Learning Environments( VLE) in order to identify which analytics techniques and sources of data accurately reflect student engagement and achievement. Methods: The interactions of 66 students with a Level 4 programming module on a VLE have been analysed via the simple K-means clustering algorithm to identify classes of behaviour and their characteristics. Results: Two prominent classes were found with students achieving higher marks attending the lectures and tutorials more regularly and accessing all types of material on the VLE more frequently than students in the lower achieving cluster. However, there were a number of exceptions that had low levels of engagement that gained high marks and vice versa. Discussion: A student’ s prior experience and characteristics of their degree programme need to be taken into account to avoid incorrectly interpreting high and low levels of engagement. Conclusions: The number of times students view online module materials will be an important factor for inclusion in any predictive LA models but must be able to take into account the differences in student backgrounds, delivery styles and subjects
Context and Objectives
Traditionally a student’ s progress and level of engagement has been measured by assessment and physical attendance. However, in a student’ s day-to-day interactions with a university, other real-time measures are being generated and stored e. g. Virtual Learning Environment( VLE) interaction, Library and Online Journal usage. The analysis of this data has been termed Learning Analytics( LA) and defined as a method for“ deciphering trends and patterns from educational big data … to further the advancement of a personalized, supportive system of higher education.”( Johnson et al., 2013). Higher Education( HE) has traditionally been inefficient in its data use( Siemens & Long, 2011) but LA has the potential to identify at-risk learners and provide intervention to assist learners in achieving success( Macfadyen & Dawson, 2010).
Examples of systems that support elements of LA include the University of Southampton’ s“ Student Dashboard”; the Open University’ s Anywhere app; the University of Bedfordshire’ s student engagement system; London South Bank University’ s partnership with IBM( Perry, 2014); Purdue University’ s Course Signals( Arnold and Pistilli, 2012) and the Student Success System( Essa & Hanan, 2012). A detailed review of systems has been published by JISC( Sclater et al., 2016), who are currently in collaboration with 50 universities to build a learning analytics service for the UK HE sector( JISC, n. d.).
In order to build a predictive Learning Analytics system, a behavioural model built from an example training set of input observations e. g. previous student VLE interaction data, is needed. However, there have been few studies that have analysed the usage of pre-existing VLEs in order to identify which analytics techniques and sources of data accurately reflect student engagement and achievement. The work presented in this paper follows on from the study by de Quincey and Stoneham( 2015) and analyses the VLE interactions of students for a Level 4 module using a clustering algorithm to identify potential groups of students with similar learning behaviour and to study the characteristics of these groups.
Methodology
The intranet within the School of Computing and Mathematical Sciences( CMS) at the University of Greenwich has been incrementally developed since 2002 and contains the key information and supports the main tasks that a student needs in order to complete their modules( 1)( Stoneham, 2012). This includes digital versions of coursework specifications, previous