基于SEM探索影响北京新疆内高班学生数学学业成绩的因素

发布时间：2018-03-29 09:22

本文选题：内地新疆高中班　切入点：学业成绩　出处：《中央民族大学》2017年硕士论文

【摘要】：内地新疆高中班自2000年办学,取得的效果有目共睹,日趋成为学者和社会关注的焦点。从2000年至今,学者对内地新疆高中班学生学业成绩的研究方法更加完善,但针对内地新疆高中班数学学业成绩的量化研究却相对较少。本论文以北京市7所内地新疆高中班学生为研究对象,采用699份高三年级测试卷以及2637份调查问卷(涵盖预科、高一、高二、高三四个年级)。首先通过分析2637份问卷得到学习效能感、学习动机和学习习惯三者的结构方程模型(structural equation modeling,SEM),然后通过验证性因子分析得到该SEM能够较好拟合高三年级的问卷,表明该结构模型具有一定的推广价值。最终结合学业成绩与三者的关系,采用不同模型间的简单比较,得到合理的交互影响模型。本论文应用SEM以及项目反应理论等多元统计分析技术,主要利用SPSS17.0、AMOS17.0以及PARSCALE4.1软件展开研究。本论文主要包括以下五个部分:第一部分,相关问题研究述评。主要采用文献分析方法,梳理国内有关研究内地新疆高中班学生数学学业成绩以及利用SEM研究学业成绩的相关文章,总结SEM在处理类似问题时的使用方法、原理以及误区。第二部分,概念界定和研究设计。对论文中相关概念以及SEM理论进行介绍,并根据SEM理论先验性特点给出参考的理论模型。同时,确定研究对象、研究工具、研究方法及研究思路。第三部分,测试卷和调查问卷的质量分析及其结构分析。首先,对高三年级测试卷进行难度、区分度、信度和结构效度分析。然后,对所有年级问卷的信度、效度进行分析,并结合相关理论框架建立合适的结构模型。最后利用高三年级问卷数据检验所建立结构模型的适配性。第四部分,探索影响学业成绩的交互模型。基于理论框架,为更好拟合所收集的数据,通过对比不同结构模型之间的相对优劣,从而确定理想的模型。第五部分,结果分析。根据理想模型找出影响学业成绩的最主要因素,得到学习效能感对学业成绩的影响最大,尤其是学习效能感中的"学生课堂专注听讲时间"对学生数学成绩的影响最为重要。最后从教师角度给出解决的建议。
[Abstract]:Since the year 2000, the high school classes in Xinjiang have achieved obvious results and become the focus of attention of scholars and society. From 2000 to now, scholars' research methods on the academic achievements of high school students in Xinjiang are more perfect. However, there are relatively few quantitative studies on mathematics achievement of high middle school in Xinjiang, China. This paper takes 7 senior middle school students in Beijing as the research object, and adopts 699 grade three test papers and 2637 questionnaires (covering preparatory courses). First of all, through the analysis of 2637 questionnaires to get a sense of learning efficacy. The structural equation modeling model of learning motivation and learning habit is obtained by confirmatory factor analysis. The results show that the structural model has certain value of popularization. Finally, combining the relationship between academic achievement and the three models, a simple comparison between different models is adopted. A reasonable interaction model is obtained. This paper mainly uses SEM and item response theory to carry out the research using SPSS 17.0AMOS17.0 and PARSCALE4.1 software. This thesis mainly includes the following five parts: the first part, Review of related issues. This paper mainly adopts the method of literature analysis to summarize the methods of using SEM in dealing with similar problems by combing the relevant articles on mathematics academic achievement of senior middle school students in Xinjiang and using SEM to study academic achievement. The second part introduces the related concepts and SEM theory in this paper, and gives the reference theoretical model according to the transcendental characteristics of SEM theory. At the same time, the research object and research tools are determined. The third part, the quality analysis and structure analysis of test paper and questionnaire. First, the difficulty, distinction, reliability and structure validity of the test paper in grade 3 are analyzed. The reliability and validity of all grade questionnaires are analyzed, and a suitable structural model is established by combining with the relevant theoretical framework. Finally, the adaptability of the structural model is established by using the data test of senior three grade questionnaire. Based on the theoretical framework, in order to better fit the collected data, by comparing the relative advantages and disadvantages of different structural models, this paper determines the ideal model. Results: according to the ideal model, the most important factors affecting academic achievement were found out, and it was concluded that learning efficacy had the greatest influence on academic achievement. In particular, the influence of "the time of listening and listening in class" in the sense of learning efficacy is most important to the students' mathematics achievement. Finally, some suggestions are given from the teacher's point of view.
【学位授予单位】：中央民族大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：G633.6

【参考文献】