Learning analytics as a tool for exploring student learning patterns

Rognoni, Bethany (2017) Learning analytics as a tool for exploring student learning patterns. Honours thesis, University of Southern Queensland. (Unpublished)

[img]
Preview
Text (Whole Thesis)
Rognoni_2017_.pdf

Download (577Kb) | Preview

Abstract

Learning analytics can provide a statistical insight into the learning behaviours
of students through the utilisation of datasets retrieved from online learning
systems (OLS). These datasets are often large and contain mixed data types,
potentially making both collation and analysis of the data complex. This
research uses demographic, assessment and OLS data from a large undergrad-
uate service course in statistics, taught at an Australian university. It provides
an exemplar of how the application of learning analytics might be performed,
using the R statistical package to implement multivariate statistical analyses.
The research focuses on both the collection and preparation of educational
data for analysis, and the application of both basic and multivariate statistical
methodologies (cluster analysis and principal components analysis) to identify
relationships between di�erent sources of data. It was found that the data
collation process is time- and resource-intensive, but valuable as the integra-
tion of di�erent data sources allows a deeper insight into the nature of student
interaction within a course. Both cluster analysis and principal components
analysis were found to provide useful interpretations of the data. The ma-
jor relationships identi�ed include: external (online) students achieve higher
grades than on-campus students; external students access OLS resources more
frequently than on-campus students; students obtain lower grades in the invig-
ilated examination than the open assignments; and students who do not access
the OLS resources tend to perform poorer on course assessments. Suggestions
for potential interventions with the aim of improving the academic perfor-
mance of students based on these trends included making early contact with
students who are not accessing course resources, and introducing an additional
invigilated assessment item to the course assessment structure.


Statistics for USQ ePrint 33948
Statistics for this ePrint Item
Item Type: Thesis (Non-Research) (Honours)
Item Status: Live Archive
Additional Information: Bachelor of Science (Honours) thesis.
Faculty / Department / School: Current - Faculty of Health, Engineering and Sciences - School of Agricultural, Computational and Environmental Sciences
Supervisors: King, Rachel; McDonald, Christine
Date Deposited: 05 Apr 2018 02:41
Last Modified: 05 Apr 2018 02:43
Uncontrolled Keywords: learning analytics; educational data; learning behaviours; online learning systems
Fields of Research : 13 Education > 1303 Specialist Studies in Education > 130306 Educational Technology and Computing
13 Education > 1303 Specialist Studies in Education > 130309 Learning Sciences
URI: http://eprints.usq.edu.au/id/eprint/33948

Actions (login required)

View Item Archive Repository Staff Only