Scanlon, Philip and Smeaton, Alan F. ORCID: 0000-0003-1028-8389 (2016) Using data analytics to predict peer-group effects on student exam results. In: Ireland International Conference on Education (IICE), 24-27 April 2016, Dublin, Ireland.
Abstract
There has been much research examining the influences on an individual student and his/her academic performance within a University environment and the impact of the heterogeneous social groups to which they become members. Manski [10] has addressed the concept as “Reflection” or the influences within group dynamics. Constructivism is a pedagogy theory that says knowledge is constructed and not acquired. Social Constructivism emphasises the importance of an individual’s social and cultural environment within which they interact and learn. It considers how they are influenced by the past, their present interactions and ergo, their influence on their peer group members. We examine this hypothesis within a University environment and build on research which recognises the intricate nature of complex community structures.
Using anonymised campus WiFi access logs collected through use of the University’s Eduroam system, we are able to identify locations where students congregate within academic and social environments and we have identified students who spend a higher proportion of their time together in comparison to with other class members, thus defining social groupings. As an illustration of our approach we randomly choose a mid-semester school day as representative of a University’s activity. We mined the 7 million WiFi log events for that day and identified the activity of 4,700 students. On that day there was an average of 40 interactions or meetings between student pairs. From this we can determine which students collocate and those who interact less with other class members
Metadata
Item Type: | Conference or Workshop Item (Other) |
---|---|
Event Type: | Conference |
Refereed: | No |
Uncontrolled Keywords: | Social Constructivism; Data Analytics |
Subjects: | Social Sciences > Adult education Computer Science > Machine learning Social Sciences > Educational technology |
DCU Faculties and Centres: | Research Initiatives and Centres > INSIGHT Centre for Data Analytics |
Funders: | This project has been funded by Science Foundation Ireland (SFI) under Grant Number SFI/12/RC/2289 |
ID Code: | 21163 |
Deposited On: | 10 Oct 2016 14:02 by Philip Scanlon . Last Modified 31 Oct 2018 11:33 |
Documents
Full text available as:
Preview |
PDF
- Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
293kB |
Microsoft PowerPoint (Conference Presentation)
2MB |
Downloads
Downloads
Downloads per month over past year
Archive Staff Only: edit this record