Information technology is deeply engrained in most aspects of everyday life and can be designed to influence users to behave in a certain way. Information technology designed for the purpose of behaviour change is referred to as persuasive systems. Examples of such systems range from e-commerce sites influencing bidding behaviour to social media encouraging the sharing of personal information. E-commerce and social media are effective at influencing broad demographics of people, which is increasingly becoming a priority for higher education as the number and types of students pursuing degrees continue to grow. The aim of this research is to construct a systematic design approach for persuasive systems in the learning environment. This is to encourage students to better engage with coursework, as this is an aspect of the learning process that is often neglected by students. Students are more concerned with assessments and obtaining good grades because these are tangible outcomes that will allow them to complete their degree. The digital learning environments students use play a key role in establishing engagement between students and course materials. Currently, learning management systems such as Blackboard Learn are the most commonly used digital learning environments. However, these do not provide strong interaction between the system and the users. To address these shortcomings, a next-generation digital learning environment, Task-Test-Monitor, was selected as the base platform and was augmented by the addition of a mobile app named Study Helper to be more persuasive. The modified Task-Test-Monitor was then used as the basis for measuring the perceived persuasiveness of such a system by students. The research was carried out over three phases, each designed to inform the next. During phase 1, two models were constructed to measure study behaviours that impacted self-perceived academic performance and behaviours that impacted a results-based measure of performance. These informed phase 2, where the persuasive system design framework was used to implement persuasive system features that supported the behaviours. The system was implemented using an agile approach over the course of two semesters, iterating features based on student feedback. Students were invited to complete a survey and participate in interviews. Phase 3 extracted persuasive factors and developed design postulates for a persuasive digital learning environment. A factor analysis was performed on the survey data and thematic and correlation analyses on the interview transcripts in order to develop the persuasive learning cycle model. The results showed that students were largely in favour of Task-Test-Monitor, with 72% wanting to use it in their other courses. Over 40% of students agreed that Task-Test-Monitor influenced them, encouraged them to complete coursework and helped them change their approach to study. The factor analysis identified three contributors to successful persuasion: guidance, visualisation and learning support. These explained over 60% of the variance in the results. The thematic analysis highlighted that learning strategy, assessment and motivation were the main concerns for students. The persuasive learning cycle model showed that students seek guidance on the work they need to do, and that visually representing their progress at the point of access of materials improves students’ motivation to continue working. Instructors play a support role by ensuring the cycle continues reliably. Persuasive learning system design postulates were extracted from the model analysis. This research has a number of implications. First, it demonstrates that a digital learning environment has the potential to be persuasive, provided it enables the persuasive learning cycle to occur. Second, instructors play an important role in the process by providing low-level support for students who require it, thus creating a scalable solution for large course enrolments. Finally, students favour a just-in-time learning model because it allows them to engage with their learning around their other pursuits and interest, which creates opportunities for persuasion to occur. Future research should test the persuasive learning cycle model in other contexts to investigate the possibility of generalising the model. The persuasive learning system design postulates should be tested as hypotheses, which will be valuable to other digital learning environments and learning system designers.
展开▼
