Personalized technology-enhanced learning (TEL) has emerged as a prominent tool used by universities to cater to students' diverse individual needs. Many higher education researchers and educators have explored the adoption of personalized TEL as an important tool to foster student learning outcomes from diverse perspectives. However, despite its significance and the substantial body of existing research, a notable gap exists in systematically evaluating the effectiveness of personalized TEL with meta-analysis approach within the higher education. To address the research gap, we investigated the effectiveness of personalized TEL in developing students' cognitive skills and non-cognitive characteristics in higher education context by utilizing the methods of meta-analysis and association rule mining. Our study reveals that the cognitive skills are reported more than non-cognitive characteristics as the learning outcomes of adopting personalized TEL. Overall, utilizing personalized TEL can improve students' cognitive skills and non-cognitive characteristics at the medium level effect size. Factors of research settings, mean of delivery, and modelled characteristics can influence students’ non-cognitive characteristics while using personalized TEL. Based on our rule mining findings, future teachers, researchers, and instructional designers can consider combining the modelling of learners' skills/knowledge or preferences with adaptive learning support strategies, such as recommending materials and scaffolding, to achieve positive effects, particularly in the fields of Social Sciences and Engineering.

中文翻译：

---

高等教育中个性化技术增强学习的有效性：具有关联规则挖掘的荟萃分析


个性化技术增强学习 （TEL） 已成为大学用来满足学生不同个人需求的重要工具。许多高等教育研究人员和教育工作者已经探索了采用个性化 TEL 作为从不同角度促进学生学习成果的重要工具。然而，尽管它的重要性和大量现有研究，但在高等教育中系统评估个性化 TEL 与荟萃分析方法的有效性方面存在显着差距。为了解决研究差距，我们利用荟萃分析和关联规则挖掘的方法，调查了个性化 TEL 在高等教育环境中发展学生认知技能和非认知特征的有效性。我们的研究表明，作为采用个性化 TEL 的学习成果，认知技能比非认知特征更多地被报告。总体而言，利用个性化 TEL 可以提高学生的认知技能和中等水平效应量的非认知特征。在使用个性化 TEL 时，研究设置、交付方式和建模特征的因素会影响学生的非认知特征。根据我们的规则挖掘结果，未来的教师、研究人员和教学设计师可以考虑将学习者技能/知识或偏好的建模与自适应学习支持策略（例如推荐材料和脚手架）相结合，以达到积极的效果， 特别是在社会科学和工程领域。