基于自我效能感的数学成绩的预测模型

Prediction Models of Academic Performance in Math Based on the Self-Efficacy

摘要: 本文以淮安市淮海中学的实际调查数据为样本进行实证研究。基于数学学习自我效能感，本文给出了数学成绩的两个预测模型：线性回归模型和Logistic回归模型。结果表明，利用加权最小二乘法得到的线性回归模型具有较高的拟合优度；而Logistic回归模型对“优”、“良”、“差”有较高的预测正确率，对“中”的预测正确率最差。

Abstract: In this paper, an empirical study is conducted by taking samples from survey in Jiangsu Huaihai senior high school. Two prediction models of academic performance in math are obtained based on the self-efficacy, which are linear regression model and Logistic regression model. The results suggest that the linear regression model obtained by weighted least square method has high goodness of fit, and the Logistic regression model has a high prediction accuracy for “excellent”, “good” and “poor”, but has the lowest prediction accuracy for “medium”.