摘要
本文通过设计调查问卷,统计学校同学对所研究因素的评分,整理问卷数据,建立以研究因素为自变量,学生平均绩点为因变量的多元线性回归模型,在建立该回归模型的过程中,发现原始变量间存在共线性,因此采用主成分分析法对研究因素降维,提取主成分,将学生成绩与提取的四个主成分做主成分回归分析。再将四个主成分与研究因素做线性回归,代入整理得到学生成绩与研究因素的最终回归方程。通过比较回归方程系数绝对值大小,可以知道在学习成绩和原始24个因素的回归表达式中,课内认真程度,课外学习时间,知识基础,阅读浏览量,社会实践经历,考前准备在所有因素中系数绝对值较大,即学习行为因子对我校学生学习程度的影响较大。
This paper designed a questionnaire, collected scores of students on the factors studied, sorted out the questionnaire data, and established a multiple linear regression model with the research factors as independent variables and students’ average score-point as dependent variables. In the process of establishing the regression model, it was found that there was collinearity among the original variables. Therefore, principal component analysis method was adopted to reduce the dimension of the research factors, extract the principal components, and perform principal component regression analysis between student achievement and the four principal components extracted. Then, the four principal components and research factors are used for linear regression, and the final regression equation of student achievement and research factors is obtained. By comparing the absolute value of the coefficient of the regression equation, it can be seen that in the regression expression of the academic performance and the original 24 factors, the absolute value of the coefficient of all factors is larger, namely, the learning behavior factor has a greater influence on the learning level of students in our school.
出处
《统计学与应用》
2023年第4期946-960,共15页
Statistical and Application
