Abstract:

Learning analytics i.e. the collection, analyses and reporting of learners’ or learning activity data, is a topic of great importance in any learning-related research. Learning analytics is used for purposes of understanding, optimizing and improving learning and the environment in which learning occurs. Basically, it could be said that the aim of LA is to get answers on “Who is doing what, how and why?” (Chatti et al., 2012) Since all parts of a learning process are connected and somehow influence one another finding the answers is a complex process that requires detailed planning and proceeding. The complexity of answers is especially present in Technology Enhanced Learning (TEL) since it is a heterogenic environment involving different participants, tools, materials and ways for its realization.

In this contribution, we focus on investigating who is doing what, by starting from the fact that students have different learning habits and that they proceed to learning in various ways i.e. they have different learning styles. As so, by who, we refer to which learning style is taking which actions. Considering learning styles and correlating it with analytics data can provide deeper understating of LA, especially in personalized learning environments, since learning styles, by definition “refer to the preferential way in which the student absorbs, processes, comprehends and retains information”.

In our work, we will rely on Felder-Silverman learning style model (FSLSM), as one of the most relevant and widely used model in determination of learners learning style in TEL. Index of learning style, a questionnaire created for identification of learning styles based on FSLSM, will be integrated and used. Further, recommendations and guides for both instructors and students will be given based on previous work and researches on FSLMS.

In this regard, we present a conceptual model of learning analytics tool for analyzing and visualizing data correlations between students’ learning styles and their behavior within an e-learning environment. The implementation is planned to be a plugin for the Learning Management System Moodle that will collect data on learners’ actions, correlate it with learners’ learning style and presents it in multiple forms of visualization.

This plugin will enable users to identify their individual learning style as well as provide additional information and recommendations for their learning habits. Using such a plugin will enable students to gain insights into their individual learning styles as well as learning styles of their peers, i.e. other students in the course or in an assigned group. On the other hand, teachers will have insight on distribution of learning styles in their courses as well as how different learning styles behave in course (e.g. which materials they use, what content they access, in which activities they participate).

While learners will benefit from actually knowing their learning style, and thus how best to approach certain kinds of information, teachers will hugely benefit from comparing individual learning styles with the average learner in their courses, and also with their own way of teaching (Felder, 1996). Thus, both the structure of the course and the used materials can be tailored to the learners’ needs, if necessary. In that way, teachers can modify their learning materials, curriculums or even their teaching style in order to create more personalized and adapted learning experience.

Keywords:

Learning styles, learning analytics, e-learning, personalized learning environment.