Year: 2008
Type: dissertation
Abstract: One of the most difficult challenges within online learning is keeping the learner motivated so as to prevent attrition or dropout. Motivation is considered to be of primary importance in determining success in online learning. Currently little research exists on intervention strategies, based on the individual learner's profile, which increase motivation in an adaptive online learning environment, so automatic intervention is not possible. The research outlined in this thesis is based on Social Cognitive Theory and the motivational constructs of self-efficacy, goal orientation, locus of control and perceived task difficulty, all of which affect the learner's self-regulation. Using systematically constructed personas, surveys were conducted with experienced classroom and online teachers to extract and validate selection rules for intervention strategies to increase learners' motivation. The results were modelled using a decision tree algorithm with correct prediction rates showing 66% to 93% accuracy. A comparison between two algorithms was carried out to predict the level of trust in the rules and in the analysis. Based on the decision rules, a motivational strategies recommender tool, MotSaRT, was designed and evaluated. The recommender tool will be incorporated into the learner model of an adaptive Intelligent Tutoring System, enabling the system to automatically select the most suitable intervention strategy to use depending on the particular student's motivational profile. This study extends the existing body of knowledge on motivational strategies by first targeting the intervention strategies to specific profiles and then by designing an automated component for the learner model of an adaptive Intelligent Tutoring System. The adaptation component will enable the system to directly intervene when a student shows signs of becoming demotivated by delivering motivation messages to the student. It is anticipated that these messages will reduce the level of attrition currently experienced in online learning.
Author: Teresa Hurley
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W791815635