Virtual Learning Lab
Gainesville, FL | August 2023 - March 2024
Principal Investigator Department
Dr. Walter Leite Education
Research Focus
The main focus of this research project was to understand how students' use of a Virtual Learning Environment (VLE) for Algebra influenced their performance on a high-stakes Algebra 1 test. The study aimed to answer key questions about the impact of different study behaviors—such as the time spent practicing, consistency in study habits, and seeking help—on students' test scores. By analyzing patterns in how students interacted with the VLE throughout the school year, the research sought to identify which behaviors were most strongly linked to academic success in Algebra. This understanding can help educators and developers improve digital learning tools and strategies to better support student achievement.
Responsibilities
In this project, I analyzed the relationships between students' use of a Virtual Learning Environment (VLE) for Algebra and their achievement on a high-stakes Algebra 1 test. I utilized log data from the VLE, applying random forests, neural networks, and clustering methods to investigate how students' time investment, study regularity, and help-seeking behaviors impacted their test performance. My tasks included feature engineering, modeling, and interpreting complex patterns in educational data. The study is specifically interested in how students used online discussion, practice questions, and video activity in a mathematics VLE. Utilizing k-means and hierarchical clustering techniques, students were classified into distinct clusters based on their interactions with the VLE. These clusters were then compared with respect to mean exam scores to assess their impact on assessment performance. The results show that students who actively engaged in making discussion posts, utilizing video controls, completing practice problems, and reviewing their incorrect responses problems, achieved better scores on a high-stakes algebra assessment. This project enhanced my skills in machine learning and educational data analysis, particularly in understanding how student behaviors in a digital environment influence academic outcomes. Through this work, I gained deeper insights into self-regulated learning and its implications for educational technology design.