摘要
预测在校大学生学业成绩并及时进行干预指导,是提升大学生学习效果的有效途径。云环境下,学习中过程性数据的积累为预测分析提供了有力依据。本研究设计并开展一门基于实体课堂和云学习平台(MOODLE平台、微信平台)的混合式大学课程。通过收集学生多类属性及学习过程性数据建立多元回归模型,对其将取得的成绩进行预测,并展开教学干预以提升其学习效果。结果显示,在预测模型方面,多元回归模型可以在全班水平上取得较佳的预测效果;在预测因子方面,绩点、在线学习参与度、前导课成绩和学习兴趣是影响学业成绩的重要因素;在干预效果方面,教学干预取得了较好效果,学弱群体接受干预后进步显著。
Predicting students' academic achievement and providing timely intervention guidance has proved to be an effective strategy for enhancing learning outcomes. In the cloud-based environment, process data generated by learning performance can be an important source of prediction. This study reports on a university course blending the physical classroom and cloud-based platforms such as Moodle and WeChat. A multiple regression model is built using various student attributes and learning process data, in order to predict academic achievement and offer timely intervention guidance. Key prediction factors include GPA, online learning engagement, preceding course achievement, and learning interest. Intervention is found to be conducive to better learning outcomes, in particular for students in difficulty.
作者
尤佳鑫
孙众
Jiaxin You Zhong Sun
出处
《中国远程教育》
CSSCI
北大核心
2016年第9期14-20,79,共7页
Chinese Journal of Distance Education
基金
国家社会科学基金重大委托项目"语言大数据挖掘与文化价值发现"(批准号:14&ZH0036)的部分研究成果
关键词
云学习平台
学业成绩
预测
干预
多元线性回归
cloud-based learning platform
academic achievement
prediction
intervention
multiple linear regression
