Modeling Disengaged Guessing Behavior in a Vocabulary Learning App using Student, Item, and Session Characteristics

Abstract

Disengagement from task content in educational apps may have a severe negative impact on learning outcomes. In the current study, we propose repeated mistakes as an indicator of disengaged guessing behavior that may be detrimental to learning. Furthermore, we propose a hierarchical generalized linear model to examine predictors of disengaged guessing behavior relating to student, item and session characteristics. Knowledge of how different characteristics contribute to the prediction of disengaged guessing may provide important information to teachers regarding which students are likely to engage in such behavior, as well as to app developers regarding the functioning of different types of task and session content.