基于广义线性模型的大学生挂科影响因素分析

The Analysis on Influence Factors of College Students’ Failing in Tests Based on Generalized Linear Model

摘要: 大学生挂科情况是学生在校学习情况的集中体现，本文通过统计建模确定了一些对大学生挂科门数具有显著影响的因素，给出了一些提高大学生在校学习情况的意见。统计建模中鉴于泊松回归模型存在过离散现象，采用了负二项回归模型。

Abstract: College students’ failing in tests represents intensively the students’ learning situation in the col-lege. This paper identified a number of factors that have a significant impact on the total number of college students’ failing in tests by statistical modeling and gave some advices that improve the college students’ academic performance. The phenomenon over discrete came out in the process of Poisson regression, so we used a negative binomial regression model.