摘要
在混合式教学过程中,教师往往通过课堂或在线测试的方式,针对学生知识掌握程度做出形成性评价,从而方便教师和学生调整教学与学习策略,提高教学效果。根据测试题的难度与区分度,建立合适的测试题库,对有效的评估学生的学习成效至关重要。文章通过二参数逻辑斯蒂模型与传统方法、单参数逻辑斯蒂模型的比较,验证了二参数逻辑斯蒂模型对于难度和区分度估计的合理性,并根据研究结果,对混合式教学中形成性评价中测试题的选择提出建议。
In the process of blended learning, teachers often make formative evaluations of students' knowledge mastery through classroom or online testing, which is convenient for teachers and students to adjust teaching and learning strategies and improve teaching effects. According to the difficulty and discrimination of the test questions, it is very important to establish a suitable test question bank for effective evaluation of students' learning outcomes. By comparing the two-parameter logistic model with the traditional method and the single-parameter logistic model, the rationality of the two-parameter logistic model for difficulty and discrimination estimation is verified. According to the research results, suggestions are made for the selection of test questions in formative evaluation in blended learning.
作者
张小娟
张睿
杜娟
Xiaojuan;ZHANG Rui;DU Juan(School of Physics Science and Engineering,Tongji University,Shanghai 200092;Shanghai Guangming High School,Shanghai 200092)
出处
《物理与工程》
2019年第2期70-73,111,共5页
Physics and Engineering
基金
上海市哲学社会科学基金项目,技术支持学习环境下的学习成效研究(项目批准号:A1610)
同济大学研究生教育改革与研究项目经费资助(2500106009)