摘要
持续改进能长期监测和提高高校的教学质量,如何基于评价结果进行科学有效的课程教学持续改进已经成为当下的热点课题。本文引入因果推断科学的“后门调整法”,分析了学情混杂控制对于教学因果推断具有的重要作用,从机理上解释了为什么需要加强包括学习动机、策略、情感、态度和价值观等多元化内容的过程性评价,提出了基于学情混杂控制的持续改进模式。首先基于多元化评价内容分析存在的学情混杂因子,构建课程教学因果图,并引入后门调整法控制混杂变量,计算教学方法对学习成果的因果效应;其次,提出消除混杂干扰的改进措施,并通过因果图分析改进的理由以及干预效果;最后,在教学中进行实施,并依据新的评价结果制定未来改进计划。实践结果表明该模式具有科学性和有效性。
Continuous improvement can monitor and improve the teaching quality of colleges and universities for a long time. How to carry out scientific and effective continuous improvement of course teaching based on the evaluation results has become a hot topic at present. This paper introduces the “back-door adjustment” of causal inference science, analyzes the importance of learning confounding control in teaching causal inference, and explains from the mechanism why it is necessary to strengthen the process evaluation of diversified contents including learning motivation, strategies, emotions,attitudes, and values, and proposes a continuous improvement model based on the control of learning confounding. Firstly, analyze the existing learning confounding factors based on diversified evaluation contents, construct a causal diagram of course teaching and control the confounding variables by back-door adjustment, to calculate the causal effect of teaching methods on the learning outcomes.Secondly, propose improvement measures to eliminate the interference of confounding, and analyze the reasons for the improvement and the effects of the intervention through the causal diagram. Finally, it is implemented in teaching, and plans for future improvement are developed based on the new evaluation results. The practical results show that the model is scientific and effective.
作者
龙璇
段斌
柯其聪
LONG Xuan;DUAN Bin;KE Qicong(School of Automation and Electronic Information,Xiangtan University,Xiangtan,Hu’nan 411105;Accreditation Guidance Center,Xiangtan University,Xiangtan,Hu’nan 411105)
出处
《电气技术》
2022年第10期67-73,共7页
Electrical Engineering
基金
湖南省学位与研究生教育教改研究重大项目(2020JGSZ016)
湖南省新工科研究与实践项目(202012)。
关键词
课程质量评价
持续改进
过程性评价
因果推断
后门调整法
course quality evaluation
continuous improvement
process evaluation
causal inference
back-door adjustment
