Back to search

Πανεπιστήμιο Μακεδονίας

Supporting learners’ capacity for autonomous decisions using learning analytics

Abstract

dc:description

This dissertation aims at supporting autonomous learning decisions using learning analytics and at introducing a model that holistically assesses the development of autonomous learning capacity. The work presented in this thesis was conducted with five overarching research objectives in mind. These objectives were: a) to deeper understand the underlying factors that justify learners’ actions in self-assessment, from an interactions analysis perspective, b) to develop coherent learner models in run-time, using the previously identified factors, in order to deliver timely personalized learning experience, c) to design and deliver task-related metacognitive help to the learners, in order to foster their autonomous decision making, d) to allow learners to practice autonomous control and assess the effect of the autonomous decisions on learning performance, and e) to explore autonomous capacity development within collaborative, group-learning conditions. This dissertation is the first attempt to employ learning analytics to understand, interpret and model the parameters related to autonomous learning capacity development. The relevant literature presents a gap in this thematic area.This thesis explores three learning analytics-enhanced approaches of guiding and supporting learners’ autonomous capacity development: (a) as controlled selection of learning tasks guided by the adaptive online learning environment, allowing the learners to practise self-regulated learning strategies in self-assessment conditions, (b) as on-demand task-related analytics visualizations targeting at enhancing learners’ autonomous metacognitive help-seeking and data-driven sense-making in online self-assessment conditions, and (c) as game-theoretic group-recommendations to motivate learners’ autonomous decision-making in collaborative learning conditions. Taken together, these three approaches provide a comprehensive and holistic view of how the online learning environment can assist learners to develop their capacity for autonomous learning. The analytics used in this thesis were extracted using different data mining techniques and cover a wide range of parameters that are associated with the learners’ behavioural, motivational and cognitive states. The overall contribution of this PhD study involves: 1) It employs learning analytics for the identification of the most informative factors that sufficiently and accurately explain performance and achievement behavior in self-assessment contexts (with multiple-choice learning tasks), in a meaningful manner, and the construction and evaluation of the respective holistic conceptual model. A corresponding model did not exist in the relevant literature; 2) The development of coherent learner models using the abovementioned factors, and enhanced with temporal dynamics for granting current-awareness, by adopting an adaptive data-stream classification approach; 3) For the first time in the literature, it develops and uses learning analytics to measure autonomous interactions during adaptive and fixed self-assessment processes of knowledge to support autonomous learning capacity development. The application of Game Theory for decision-making and guiding group recommendations is considered original, as well. Finally, the introduction of an analytics-driven model for assessing autonomous learning capacity development in online self-assessment conditions, based on self-regulated learning perceptions and learning analytics that practically record the self-regulated learning interactions is an innovative contribution that opens a new research area.

Degree

thesis:*
Grantor dc:publisher
Πανεπιστήμιο Μακεδονίας
Year dc:date
2018

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Papamitsiou, Zacharoula
  • Παπαμήτσιου, Ζαχαρούλα

Subjects

dc:subject × 20

Rights

dc:rights
Statement dc:rights
  • BY_NC
Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/44096

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Papamitsiou, Zacharoula; Παπαμήτσιου, Ζαχαρούλα. Supporting learners’ capacity for autonomous decisions using learning analytics. Πανεπιστήμιο Μακεδονίας, 2018. http://hdl.handle.net/10442/hedi/44096