Learning Analytics Dashboards for Assessing Remote Labs Users' Work: A Case Study with VISIR-DB

Abstract

AbstractIn science and engineering education, remote laboratories are designed to bring ubiquity to experimental scenarios, by having real laboratories operated through the Internet. Despite that remote laboratories enable the collection of students' work data, the educational use of these data is still underdeveloped. Learning analytics dashboards are common tools to present and analyze educational data to provide indicators to understand learning processes. This paper presents how data from remote labs, such as Virtual Instruments Systems In Reality (VISIR), can be analyzed through a learning analytics dashboard to help instructors provide better feedback to their pupils. Visualizations to study the use of the VISIR, to assess students’ performance in a particular activity and to facilitate the assisted assessment of students are introduced to the VISIR dashboard (VISIR-DB). These visualizations include a new recodification of circuits that keeps the fragment being measured, in order to better identify student’s intention. VISIR-DB also incorporates functions to check a priori steps in the resolution process and/or potential errors (observation items), and logical combinations of them to grade students' performance according to the expected outcomes (assessment milestones). Both work indicators, observation items and assessment milestones, can be defined in activity-specific text files and allow for checking the circuit as coded by the interface, the conceptual circuit it represents, its components, parameters, and measurement result. Main results in the use of VISIR for learning DC circuits course show that students mainly use VISIR when indicated by instructors and a great variability regarding to time of use and number of experiments performed. For the particular assessment activity, VISIR-DB helps to easily detect that there is a significant number of students that did not achieved any of the expected tasks. Additionally, it helps to identify students that still make a huge number of errors at the end of the course. Appropriate interventions can be taken from here.