基于多元统计分析的学生综合素质评价

发布时间：2018-05-29 13:01

本文选题：学生综合素质评价 + 多元统计分析　；参考：《清华大学》2013年硕士论文

【摘要】：当前，高校学生的综合素质越来越成为关注的热点，对此，科学的评价系统也就很重要和必要。然而，目前大多高校采取的以学习成绩加权平均的传统评价系统已经暴露出很多问题，因而如何建立科学的综合素质评价体系就成为素质教育过程中亟待解决的问题。本文采用多元统计的模型对大学生综合素质进行测评，主要运用了主成分分析，因子分析及聚类分析的方法，在力保信息丢失最少的原则下，降低多变量数据系统的维度，提高数据信息的分析效率。同时,本文对主成分分析法进行改进，提出两种改进的方法，并与传统的主成分分析法进行比较研究。结合当前高校课程和学生活动的设置，本文将学生综合素质按照学业成绩、社会活动、专业技能等分为9个指标，，采用上述模型，利用SPSS和MATLAB软件，将原来各项素质指标组合成一组新的较少的综合指标，尽可能多的保留原始信息，为每位同学做出较为全面、客观、合理的评价。该方法克服了当前各高校采用的成绩加权平均对学生进行笼统排队的弊端，能够全面综合的对大学生综合素质进行评价，具有良好的使用性和可操作性，并且能具体的体现每位学生的优势和特点。通过实例分析，本文最后又进行了总结，提出模型中不完善之处，并提出一些展望。
[Abstract]:At present, the comprehensive quality of college students is becoming more and more concerned, so the scientific evaluation system is very important and necessary. However, many problems have been exposed in the traditional evaluation system based on weighted average of academic achievement, so how to establish a scientific comprehensive quality evaluation system has become an urgent problem in the process of quality education. This paper uses the multivariate statistical model to evaluate the comprehensive quality of college students. It mainly uses the methods of principal component analysis, factor analysis and cluster analysis to reduce the dimension of multivariable data system under the principle of keeping the information lost least. Improve the efficiency of data analysis. At the same time, the principal component analysis method is improved, and two improved methods are put forward, and compared with the traditional principal component analysis method. Combined with the current curriculum and student activities in colleges and universities, the comprehensive quality of students is divided into 9 indexes according to academic achievement, social activities and professional skills. The above model is adopted, and the software SPSS and MATLAB are used. The original quality indicators are combined into a group of new and less comprehensive indicators, as much as possible to retain the original information, for each student to make a more comprehensive, objective, reasonable evaluation. This method overcomes the disadvantage of the weighted average of achievement in colleges and universities, and it can evaluate the comprehensive quality of college students comprehensively, and has good usability and maneuverability. And can embody each student's superiority and characteristic concretely. Through the analysis of examples, this paper summarizes the imperfections of the model, and puts forward some prospects.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：G645.5

【参考文献】