Scholarlink

<p>Traditional distance learning evaluations often fail to diagnose students’ deficiencies in the early stages. This research tackles this challenge by implementing online formative assessments and analyzing the records of an introductory physics course, aiming to predict at-risk students. These online assessments go beyond mere evaluation, offering valuable benefits for students, such as immediate feedback, the opportunity to retake assessments, and learn from mistakes, all fostering deeper understanding. Three machine learning algorithms were used to predict students’ final situation. All algorithms demonstrated good classification performance. However, the support vector machine (SVM) algorithm surpassed the others. This result allows us to predict potential failure during the very first formative assessment. This method empowers instructors to intervene early and improve student success, potentially leading to higher retention rates. These findings pave the way for personalized learning interventions in distance learning education. Potentially transforming students’ outcomes and fostering a more engaging learning experience.</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

Early Detection of At-Risk Students in Physics Through Remediated Online Formative Assessments and Machine Learning

  • 期刊论文

  • 发表日期：

    2025年6月2日

查看全文
  • Charlie V. Sarmiento,
  • Germano Maioli Penello,
  • Lucas Sigaud

摘要

Traditional distance learning evaluations often fail to diagnose students’ deficiencies in the early stages. This research tackles this challenge by implementing online formative assessments and analyzing the records of an introductory physics course, aiming to predict at-risk students. These online assessments go beyond mere evaluation, offering valuable benefits for students, such as immediate feedback, the opportunity to retake assessments, and learn from mistakes, all fostering deeper understanding. Three machine learning algorithms were used to predict students’ final situation. All algorithms demonstrated good classification performance. However, the support vector machine (SVM) algorithm surpassed the others. This result allows us to predict potential failure during the very first formative assessment. This method empowers instructors to intervene early and improve student success, potentially leading to higher retention rates. These findings pave the way for personalized learning interventions in distance learning education. Potentially transforming students’ outcomes and fostering a more engaging learning experience.