Hello everyone. This is Xue-Wang Geng, a third-year doctoral student.

I would like to introduce a paper I read for our English seminar. Below is a summary of the content described in the paper.

Paper Title: Adaptive or adapted to: Sequence and reflexive thematic analysis to understand learners’ self-regulated learning in an adaptive learning analytics dashboard
Journal: British Journal of Educational Technology
Pages: 1-28
Publication Year: 2022
Authors: Eunsung Park, Dirk Ifenthaler, Roy B. Clariana

With the development of ICT, there is an increasing number of solutions that support Self-Regulated Learning. Among these, Learning Analytics (LA) is one such solution. LA is defined as the measurement, collection, analysis, and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs (Siemens & Long, 2011). It has been reported that using LA enables:
・Support for controlling and monitoring learning situations
・Promotion of reflection and awareness of learning engagement
・Improvement of motivation, learning experience, and retention rates
and so on.

However, the authors point out that many LA studies focus on promoting learning performance and predictive models, and why using LA is effective has not been sufficiently investigated. Therefore, it is necessary to clarify how learners change behaviorally and strategically in their Self-Regulated Learning through interventions utilizing LA. Furthermore, they argue that we should grasp how learning takes place by using LA from the learners’ perspective. In the paper I selected, the authors designed an Adaptive Learning Analytics Dashboard (ALAD) and focused on how learners start and conduct their learning with the ALAD, investigating learners’ learning experiences and perceptions when interacting with the ALAD.

Regarding the ALAD in this study, a warm-up test (WU, a pre-knowledge test) was provided at the beginning of each lesson. After learners took the WU test, the test scores were reflected in their individual ALAD. In the ALAD, the number of contents accessible to learners differs depending on whether they take the WU test and their test scores. Furthermore, learners can check information from the ALAD such as their current learning status, time spent, and the remaining time until completing each lesson. Additionally, the ALAD suggests the next content that learners should study.

To investigate how learners were studying in the ALAD, an experiment was conducted over one semester with 81 students taking a finance course. The course consisted of two face-to-face classes per week and online learning via the ALAD. During the implementation period, all interactions between learners and the ALAD were recorded. Also, interviews were conducted with 12 learners at two points in time.

As a result, three groups classified by cluster analysis regarding the use of the WU test were identified: the WU group, which mainly used the WU test; the content group, which focused on learning content; and the mixed group, which balanced the WU test and content learning more evenly in terms of time. Then, as a result of sequence analysis, differences were recognized in the learning sequences of the three groups. Learners in the WU group took the WU test at the beginning of each lesson and then studied the content, whereas the content group skipped the WU test and spent time learning the content. Furthermore, the results of the reflexive thematic analysis provided insights into how learners used the ALAD functions to construct learning strategies and how they monitored and controlled their learning. In particular, it is suggested that different ALAD usage arose from differences in awareness, such as whether the WU test is useful for achieving learning goals. For example, the WU group began to use the test as a means to achieve their goals (high scores) and constructed learning strategies to make the most of the test. In contrast, the content group recognized that the test was not useful for achieving their goals and employed a different learning strategy of proceeding with learning content without taking the test. Note that this paper proposes several considerations for dashboard design.
・Recommendation functions should be designed by instructors or teachers in advance based on what learners need during their learning, rather than data-driven suggestions based on whether the test answers are correct or incorrect.
・If a test can be taken multiple times, it is necessary to provide learners with additional and alternative resources for the test to avoid meaningless trial and error.

As for my impressions, the qualitative evaluation part of this paper, which uses reflexive thematic analysis to examine the impact of dashboard usage on learners’ learning behaviors and strategies, was very helpful as an analysis method for the research I am currently working on. However, while this paper discussed how to use a dashboard to support Self-Regulated Learning as a whole, I am curious about how the dashboard was designed to facilitate Self-Regulated Learning. Also, I think it would be more interesting if there were considerations based on specific learning behaviors. For example, I felt it was necessary to investigate what information learners check on the dashboard before proceeding to the next content learning or WU test. In my future research, I would like to explore the impact of dashboard functions on learning behaviors at a finer granularity.