基于Adaptive-Lasso Logistics模型的方剂学学习动机影响因素分析

目的讨论Adaptive lasso-logistic回归模型在方剂学学习动机的影响因素中的应用。方法基于成都市某大学在校2017级学生进行分层抽样,综合运用方差分析及lasso-logistic回归模型,利用R软件分析学生在方剂学学习过程中关于学习动机的影响因素。结果模型共纳入662名学生,通过交叉验证法,可知Adaptive lasso-logistic回归模型中纳入的变量为:外驱动机、学习信念、考试焦虑、性别、专业、科别。从时间成本和模型简洁度来看,Adaptive-lasso logistic模型优于全变量logistic模型和ridge logistic模型。结论Adaptive-Lasso Logistics回归模型在压缩模型、变量筛选以及预测准确率上更有效,有助于了解学生的学习动机,为教学工作者后续教学安排提供一定的参考意见。

Analysis of Influencing Factors of Learning Motivation of Formulaology Based on the Adaptive-Lasso Logistics Model

Objective To discuss the application of Adaptive lasso-logistic regression model in the influencing factors of learning motivation of formulaology.Methods Based on stratified sampling of students in 2017 at a university in Chengdu,a comprehensive analysis of variance and lasso-logistic regression models were used.R software was used to analyse the influencing factors of students'learning motivation in the process of formulaology learning.Results A total of 662 students were included in the model.Through cross-validation,the variables included in the Adaptive lasso-logistic regression model were external drive,learning belief,test anxiety,gender,major,and department.From the perspective of time cost and model simplicity,the Adaptive-lasso logistic model is superior to the full-variant logistic model and the ridge logistic model.Conclusion The Adaptive-Lasso Logistics regression model is more effective in compression model,variable screening and prediction accuracy.It is helpful to understand students'learning motivation and provide some reference for the follow-up teaching arrangement of educators.