鲁东大学学报(自然科学版)

为进一步提高高校生源质量，本文选取可能影响生源质量的多个因素，并从权威数据网站上获取所需数据用于研究分析。首先对高校生源质量与各因素作相关性分析，并应用SPSS软件建立多元线性回归模型；其次，对模型进行显著性检验，并运用逐步回归法进行变量选择，最终得到“最优”多元线性回归模型。基于上述数据，运用Lasso回归进行变量选择，检验经过筛选处理后变量的显著性，最终确定对高校生源质量有显著影响的因素。本文所用两种方法得到的结果均表明：本科生人数、重点学科、综合声誉指数以及高校是否属于理工类学校对高校生生源质量具有显著影响。

In order to further improve the quality of college students, several factors that may affect the quality of college students were selected in this paper,and the required data was obtained from the authoritative data website for research and analysis.Firstly, the correlation between the quality of college students and various factors was analyzed, and a multiple linear regression model was established by using SPSS software.Secondly, the model was tested for significance, and the stepwise regression method was used to select variables, and finally the optimal multiple linear regression model was obtained.Based on the above data, Lasso regression was used to select variables to test the significance of the variables after screening, and finally determine the factors that have a significant impact on the quality of college students.The results obtained by the two methods in this paper show that the number of undergraduates, key disciplines, comprehensive reputation index and whether a university is a science and engineering school have significant impacts on the quality of college students.