多模态数据赋彩学习投入测评新发展

New Development of Multi-modal Data Color Learning Investment Evaluation

学习投入作为衡量学习者成功与否的重要指标、影响学习者学习绩效的关键因素,受到专家学者的广泛关注。如何对学习投入进行合理有效测评,是当前教育领域不可回避的现实课题。通过对传统测评方式梳理发现其极易产生“路灯效应”,存在分析信度及效度不准确等弊端。本研究以多模态数据为切入点,提出多模态数据赋彩学习投入测评新机制,包括多模态数据信息采集、关系挖掘、融合应用三个层面,多维解析学习投入测评逻辑机理,助力突破学习投入测评现实困境,以推动学习投入测评新发展。

As an important indicator to measure learners'success and a key factor affecting learners'learning performance,learning engagement has attracted extensive attention of experts and scholars.How to evaluate learning investment reasonably and effectively is an unavoidable practical topic in the field of education.By combing the traditional evaluation methods,it is found that it is very easy to produce"street lamp effect",which has some disadvantages,such as inaccurate analysis reliability and validity.Taking multimodal data as the starting point,the research puts forward a new mechanism for the evaluation of learning investment with multimodal data,including three levels of multimodal data information collection,relationship mining and fusion application,and multi-dimensional analysis of the logical mechanism of learning investment evaluation,which help breaking through the practical difficulties of learning investment evaluation so as to promote the new development of learning investment evaluation.