Scholarlink

<p>Educational data mining techniques were used to analyze the 2022 Program for International Student Assessment (PISA) dataset to identify key predictors of mathematics and science achievement among Canadian students. Using models including support vector regressor (SVR), random forest (RF), and least absolute shrinkage and selection operator (LASSO), the study reveals that mathematics self-efficacy, home possession, curiosity, and mathematics anxiety influence student achievement in mathematics and science. These factors are in line with the ecological systems theory utilized in this study, emphasizing the significance of both individual competencies and environmental factors in educational achievement. Random forest emerged as the model with the highest accuracy, thereby highlighting its efficiency in educational research. The findings suggest targeted interventions that could enhance science, technology, engineering, and mathematics (STEM) education, informing both educational practices and policy decisions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Investigating Predictors of Student Performance in STEM Using Educational Data Mining Techniques

  • 期刊论文

  • 发表日期：

    2025年6月1日

查看全文
  • Doris Abroampah,
  • Ying Cui,
  • Mark Gierl

摘要

Educational data mining techniques were used to analyze the 2022 Program for International Student Assessment (PISA) dataset to identify key predictors of mathematics and science achievement among Canadian students. Using models including support vector regressor (SVR), random forest (RF), and least absolute shrinkage and selection operator (LASSO), the study reveals that mathematics self-efficacy, home possession, curiosity, and mathematics anxiety influence student achievement in mathematics and science. These factors are in line with the ecological systems theory utilized in this study, emphasizing the significance of both individual competencies and environmental factors in educational achievement. Random forest emerged as the model with the highest accuracy, thereby highlighting its efficiency in educational research. The findings suggest targeted interventions that could enhance science, technology, engineering, and mathematics (STEM) education, informing both educational practices and policy decisions.