Modeling Student Growth Over Time in High School Algebra I from a Digital Learning Environment

Algebra 1 holds significant importance as a mathematics course due to its content and its potential impact on students' long-term outcomes. Thus, monitoring students' achievement in Algebra 1 becomes crucial. Digital learning products have made tracking such progress more accessible, but there is a limited body of research exploring the distinct features of digital learning product data and their influence on measurement. This study aims to address this gap by analyzing item response data obtained from a K-12 learning platform to understand students’ growth in Algebra 1, considering the unique characteristics of response patterns.

Mixed-effect logistic regression models were employed to estimate students’ initial abilities and growth. Additional specifications were implemented to control for item difficulties and address extended response gaps. The findings of the models suggested that an average student's learning rate aligns with the rate of increase in item difficulties. The study also discussed its limitations and suggestions for data quality enhancement.