< Back to previous page

Project

Designing and evaluating theory-informed, aggregated and actionable learning analytics dashboards

The topic of this project is the design and evaluation of Learning Analytics Dashboards (LADs). Despite the vast amount of data that is captured in digital learning applications, their visualizations in LADs currently remain underused within the teaching and learning process. To improve the overall effectiveness of digital learning tools, it might be important to visualize data (a) that relate to indicators with a theoretically and empirically proven relevance to represent the learning process, (b) that are based on learners’ activities in different applications, and (c) that are formulated in actionable terms. With teachers’ data literacy skills often being limited, such LADs can help them to get insights into students’ learning processes and consequently act upon the provided information. The project is related to AI in education as the algorithms underlying LADs require advanced explainable AI techniques to transform raw data into easily interpretable visualizations. In view of increasing the effectiveness of educational technology tools there has been a growing attention to the development of LADs that can provide teachers insight in students’ learning processes. However, EdTech companies lack knowledge about which learning parameters to measure, how to accurately measure them and how to meaningfully convey them to teachers. Research has rarely focused on how learning theories can inform the design of LADs. More particularly, in the current project, we will rely on theories regarding self-regulation and self-determination to determine which indicators can be visualized on LADs and how to present them in an actionable way. In this PhD we will investigate which indicators can be most informative to be presented on learning analytics dashboards. We will focus on theory- and evidence-informed indicators that are visualized as actionable data. These indicators will be based on aggregated data from different technological tools that are used for learning. Actionability refers to presenting transparent information that can be acted upon to ultimately lead to more personalization of the learning process. The impact of LADs on teachers’ instructional practices will be evaluated. This project aims to improve LADs by substantiating them with theories and evidence on learning and instruction, by making them more transparent and easier to understand and interpret, and by providing trustworthy recommendations for follow-up actions. As such, this project will lead to better support for instructors to further personalize the learning process.

Date:1 Feb 2024 →  Today
Keywords:learning analytics dashboards
Disciplines:Educational technology
Project type:PhD project