Lesson Time and Accuracy in a Computer-Delivered Reading Intervention

Despite extensive national efforts to improve literacy outcomes, the United States continues to face a significant reading crisis, particularly among upper elementary and middle school students. While early interventions often focus on word reading difficulties in young children, older struggling readers face more complex challenges, such as deficits in fluency and comprehension, yet receive limited targeted support. In recent years, educational technology has emerged as a promising tool to deliver targeted, individualized reading interventions that promote student engagement and alleviate instructional demands on teachers. This study explores the relationship between time-on-task—a proxy for student engagement—and comprehension accuracy within a computer-delivered reading intervention for struggling middle school readers. Specifically, the study examines whether lesson completion time predicts accuracy in inference-making tasks and investigates how students' response patterns (e.g., fast and accurate, slow and inaccurate) align with their performance. Grounded in the speed-accuracy trade-off framework, which posits a balance between speed and precision in task performance, the study seeks to expand the understanding of how older students interact with digital reading tools. By analyzing both completion times and accuracy across lessons, this research aims to inform the design of more effective computer-based interventions tailored to the diverse needs of struggling adolescent readers. Results revealed no significant correlation between lesson speed and accuracy (t = .041, p = .784, r = -.019) suggesting that time-on-task did not significantly influence reading comprehension outcomes in this sample. Student response pattern types highlight variability in engagement and performance within computer-delivered reading intervention.