Temporal Trends In Textbook Tracking Data

Project introduction and background information

Abstract

Learning analytics can be a very useful tool to improve the design of our courses.  We used the Living Textbook, a concept-based digital textbook, to collect tracking data during an introductory MSc course. Via the reading analytics we identified the optimal length of learning paths, the optimal number of words in concept descriptions and other information that can improve our courses. 

Project Introduction and background information

Learning analytics are widely used in higher education to monitor and predict student success, and for learning design purposes. This paper explores the use of textbook tracking data, collected during an  introductory course on Geographic Information Science (GIS). The digital textbook, called the Living Textbook (LTB), is a combination of a concept map and a wiki with additional learning functionalities, including sequential learning. Learning pathways are chains of concepts addressing a single learning objective that lead students through the reading assignments. They are comparable with a chapter in a regular book. 

Objective and expected outcomes

The learning behaviour of students is often non-linear and does not always follow the predefined learning pathways due to difference in learning experience and learning styles. The same user can show different reading behaviour during different phases of the course. This study adds the perspective of time into the analysis of tracking data. The question addressed in this paper is: “Which temporal patterns can we find in students’ use of the Living Textbook, and how can we use this information to optimize the learning pathways?”

Results and learnings

The total number of visits per day of the course is a very misleading number to base conclusions on. We see large peaks before the exam, but many of these concept visits were re-visits by students that earlier studied these concepts. The later a learning pathway is scheduled in the course, the more first-time concept visits are generated close to the exam. For early teaching, there is enough time to review the materials. Teachers should take this into account when designing their courses. The same applies to the order of concepts in the learning pathways. The earlier a concept is positioned in this learning pathway, the longer the time spent by students in studying these topics, especially for subsequent visits. Further analysis can help to determine the preferred length of the learning pathways.

Practical outcomes

This research has also been presented at the 48th SEFI Annual Conference on Engineering Education, SEFI 2020, Online, Enschede, Netherlands (20 Sep 2020 - 24 Sep 2020). More could be found in the paper (at the downloads on the right): “Temporal trends in textbook tracking data.” by P.W.M. Augustijn-Beckers, M.J. Verkroost and I.Oliveira

Attachments 

Paper: Temporal trends in textbook tracking data.” by P.W.M. Augustijn-Beckers, M.J. Verkroost and I. Oliveira