Rapid Guessing Behavior Detection in Microlearning: Insights Into Student Performance, Engagement, and Response Accuracy

This study investigates rapid guessing behavior (RGB) in a microlearning environment, which has gained importance due to its growing user base in formal and informal education. The presence of RGB in question responses can distort the perceived difficulty of the content and skew the overall assessment of a student’s abilities. When test-takers respond without careful consideration, their scores may not accurately reflect their knowledge and skills, potentially leading to incorrect conclusions about student performance and the effectiveness of the educational content. The primary goal of this research is to evaluate whether methods designed for RGB detection in traditional testing environments can be effectively applied in microlearning-a low-stakes context that contrasts with traditional high-stakes testing scenarios. Three different approaches to RGB identification, represented by six methods, were selected for this study. These include the nominal time method based on average response time (NT10), visual identification of bimodal distributions in response times, and a group of methods that combine response time and accuracy (RTRA and CUMP). Additionally, the study aims to identify variables that can categorize students based on their behavior and performance and examine how these variables can improve the accuracy of RGB detection. Analyzing real-world data, the study focuses on student interactions with microlearning content, particularly regarding response times, accuracy, and engagement levels. The findings confirm that methods developed for RGB identification in traditional settings can be adapted for use in microlearning environments. Furthermore, the study reveals significant differences in RGB identification when students are clustered by reading speed and learning performance.