Learning Analytics and Computerized Formative Assessments: An Application of Dijkstra’s Shortest Path Algorithm for Personalized Test Scheduling

The use of computerized formative assessments in K–12 classrooms has yielded valuable data that can be utilized by learning analytics (LA) systems to produce actionable insights for teachers and other school-based professionals. For example, LA utilizing used monitoring students’ progress reading identifying struggling readers. Using such systems, also determine whether is adequate as the student works towards their instructional goal. However, due lack guidelines on timing, number, frequency assessments, often follow a one-size-fits-all approach testing all students together pre-determined dates. This leads rigid test scheduling ignores pace at which improve skills. In some cases, consequence yields little no useful data, while increasing amount time miss. this study, we propose an intelligent recommender system (IRS) based Dijkstra’s shortest path algorithm optimal assessment schedule each throughout school year. We demonstrated feasibility IRS using real from large sample grade two (n = 668,324) four 727,147) who participated series assessments. Also, conducted Monte Carlo simulation study evaluate performance presence unusual growth trajectories (e.g., negative growth, plateau). Our results showed could reduce number administrations required both levels eliminating did not change substantially. addition, indicated yield robust with meaningful recommendations under relatively extreme trajectories. Implications education future research are discussed.