One of the challenges for the future of technology-enhanced learning is the retention of learners. On-line learning environments should engage learners and provide an appropriate “Quality of Experience” (QoE). For more than a decade, adaptive hypermedia systems have been used to adapt content and instruction to individual knowledge, goals and preferences in an effort to engage learners. However, even if the content is highly engaging it can be very difficult to achieve good Quality of Experience for people without sufficient technological infrastructure or fixed access, such as in rural or remote areas and learners in developing economies. Despite recent improvements in network technology, any increases in bandwidth are quickly consumed by more demanding applications. So far QoE has been approached primarily from an engineering perspective, considering technical factors such as download times or video quality. But arguably QoE in multimedia e-learning systems must be viewed in a broader sense and has to be aligned with the concept of user experience to capture more dimensions of the experience. In this context this thesis proposes a new model for QoE in adaptive multimedia e-learning, combining QoS, learning theory and flow experience. The proposed adaptation policies for QoE combine multimedia e-learning theory with network QoS adaptations. A novel measurement model for QoE in adaptive multimedia elearning was developed to measure QoE and the factors influencing QoE and to explore the inter-relationship of various aspects of QoE. This research identified learning and flow experience as key contributors to QoE in multimedia e-learning systems. Moreover, the results indicate that an international Delphi expert study under-estimated the impact of Quality of Service on QoE in multimedia e-learning; the research reported in this thesis, including network simulations and extensive user tests, showed that the impact of QoS deserves more consideration. The proposed QAMM2 algorithm combining QoS adaptation and media mix adaptation was the subject of simulation and extensive user testing which demonstrated that QoE in multimedia e-learning systems can indeed be improved by adjusting the content media format to suit network conditions.