The Open UniversitySkip to content
 

Developing Learning Analytics for Epistemic Commitments in a Collaborative Information Seeking Environment

Knight, Simon James Goodwin (2016). Developing Learning Analytics for Epistemic Commitments in a Collaborative Information Seeking Environment. PhD thesis The Open University.

Full text available as:
[img]
Preview
PDF (Version of Record) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (4MB) | Preview
Google Scholar: Look up in Google Scholar

Abstract

Learning analytics sits at the confluence of learning, information, and computer sciences. Using a distinctive account of learning analytics as a form of assessment, I first argue for its potential in pedagogically motivated learning design, suggesting a particular construct – epistemic cognition in literacy contexts – to probe using learning analytics. I argue for a recasting of epistemic cognition as ‘epistemic commitments’ in collaborative information tasks drawing a novel alignment between information seeking and multiple document processing (MDP) models, with empirical and theoretical grounding given for a focus on collaboration and dialogue in such activities. Thus, epistemic commitments are seen in the ways students seek, select, and integrate claims from multiple sources, and the ways in which their collaborative dialogue is brought to bear in this activity. Accordingly, the empirical element of the thesis develops two pedagogically grounded literacy based tasks: a MDP task, in which pre-selected documents were provided to students; and a collaborative information seeking task (CIS), in which students could search the web. These tasks were deployed at scale (n > 500) and involved writing an evaluative review, followed by a pedagogically supported peer assessment task. Assessment outcomes were analysed in the context of a new epistemic commitments-oriented set of trace data, and psychometric data regarding the participants’ epistemic cognition. Demonstrating the value of the methodological and conceptual approach taken, qualitative analyses indicate clear epistemic activity, and stark differences in behaviour between groups, the complexity of which is challenging to model computationally. Despite this complexity, quantitative analyses indicate that up to 30% of variance in output scores can be modelled using behavioural indicators. The explanatory potential of behaviourally-oriented models of epistemic commitments grounded in tool-interaction and collaborative dialogue is demonstrated. The thesis provides an exemplification of theoretically positioned analytic development, drawing on interdisciplinary literatures in addressing complex learning contexts.

Item Type: Thesis (PhD)
Copyright Holders: 2016 Simon James Goodwin Knight
Keywords: data mining; natural language processing; digital media
Academic Unit/School: Faculty of Science, Technology, Engineering and Mathematics (STEM) > Knowledge Media Institute (KMi)
Item ID: 46143
Depositing User: Ann McAloon
Date Deposited: 03 May 2016 13:35
Last Modified: 04 Oct 2016 20:26
URI: http://oro.open.ac.uk/id/eprint/46143
Share this page:

Download history for this item

These details should be considered as only a guide to the number of downloads performed manually. Algorithmic methods have been applied in an attempt to remove automated downloads from the displayed statistics but no guarantee can be made as to the accuracy of the figures.

Actions (login may be required)

Policies | Disclaimer

© The Open University   + 44 (0)870 333 4340   general-enquiries@open.ac.uk