My publishing experience the last 9 years across legal, medical education, and higher education publishing has given me product insight and experience in seeing how powerful adaptive learning platforms can be when the user is trainied adequately and time is spent on understanding various analytics in the reports, then taking action with data driven decision making to enhance and improve learner outcomes. I have often heard from professional and student users I have worked with that the reports are very time consuming to produce and understand, let alone set aside time to act on the dashboard results provided for both educator and learner. 

 After the lecture we had in educational data mining with Luc, the innovative learning practice I chose to research and discuss is the design of dashboard visualizations on learning management systems. Sedrakayan states the lack of information available on designing Learning Analytics Dashboard's with proven theories in learning science (Sedrakayan, 2019). Online learning is increasingly becoming more popular, and global learner opportunities are rising due to recent technological advancements. According to Bodily, "learning analytics 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" (Bodily, 2017). Students are often not involved with decision making based on learning management system data, as administrators and instructors leave the actual users out of the process (Bodily, 2017). In addition,  Schwendimann states that, "once data has been gathered, data needs to be processed, analyzed and visualized." (Schwendimann, 2016). Teachers lack the academic and professional training in understanding data analytics, thus it's easier and more efficient if more visualizations were available for both students and teacher to understand current insights. Andiola states that "big data is widely viewed as the next frontier for innovation, competition, and productivity..." (Andiola, 2020). The below video is an overview on how Zoola Analytics works. Zoola states that it's "...your one-stop solution that complements Moodle to effectively measure every single aspect of the learning process so that you can increase learner engagement and skyrocket training effectiveness." (Zoola Analytics, 2020).

H. Aldowah describes visual data mining as combining "...traditional data mining methods with data visualization tools in order to visualize patterns of interest...in higher education, used to graphically reduce complex and multidimensional student tracking data...to effectively analyze...the learning process" (Aldowah, 2019). By further implementing more diverse visual analytics for teachers and students, less effort can be attributed to understanding and analyzing large data sets (Vaitsis et al, 2014) and more effort can then be placed on replacing current resources and or teaching techniques and finding other ways to supplement learning. Teachers can also readily apply the effective agile methodology in their curriculum, if they efficiently use visual learning analytics. The agile methodology and framework was initially started with software engineers. Over the years, their framework has been introduced and seen across various disciplines. Part of their workings is constant iteration based on data and feedback, something the education sector can start implementing more of.  According to Confrey, an agile curriculum framework suppports adjustments that can be made by analyzing big data, and educators can further iterate their lesson planning by constantly reviewing, interpreting and acting on the data, throughout teaching and after various assessments (Confrey, 2018). 

J. Grann and D. Bushway, “Competency Map: Visualizing Student Learning to Promote Student Success,” Proc. Fourth Int'l Conf. Learning Analytics and Knowledge (LAK14), 2014; http://doi.org/10.1145/2567574.2567622

As more teachers educate and familiarize themselves with learning analytics and data mining, all the literature reviews thus far have shown positive student outcomes. Aldowah states that "...data mining techniques can provide educational policy makers with data-based models essential for supporting...goals to enhance the efficieny and quality of teaching and learning." (Aldowah, 2019).  Previously, with regards to collecting student data, the reviews and studies only focused on the learning management system logs obtained from the university (Schwendimann et al., 2016). Students previoulsy did not have the option to ask themselves "How can I do better", it was common only to see performance data (Sedrakyan, et al., 2016). 

The purpose of this case study is to examine the literature on designing learning analtytics with learning theories.

Sedrakyan, G., et al., Linking learning behavior analytics and learning science concepts: Designing a learning analytics dashboard for feedback to support learning regulation, Computers in Human Behavior (2018), https://doi.org/10.1016/ j.chb.2018.05.004

When thinking about designing effective and useful dashboards for learning analytics, the learning science concepts that Sedrakyan has included in his research are feedback, reguated learning, goal orientation, effectiveness of learning, efficiency of learning, user perference and emotion regulation. The subcategories and related topics within the feedback concepts include cognitive, behavioral, outcome-oriented, and process oriented. Within regulated learning, he includes self-regulation, co-regulation, socially shared regulation with a focus on peer and group planning, monitoring and adapting. With goal orientation, the related concepts include performance orientation, mastery orientation, performance approach avoidance and mastery approach avoidance, coupled with self regulation. The linked learning analytics dashboard concepts and learning science concepts include a self-oriented profile, learning goals, dependencies, group oriented planning profiles, user feedback and interactivity to name a few (Sedrakyan, et al 2019). 

Sedrakyan, G., Mannens, E., & Verbert, K. (2019). Guiding the choice of learning dashboard visualizations: Linking dashboard design and data visualization concepts. Journal of Computer Languages, 50, 19–38. https://doi.org/10.1016/j.jvlc.2018.11.002

How do we determine which metrics to include that will quantify results? Sedrakyan, et al (2019) states that effectiveness, expressiveness, readbility, and interactivity should all be considered, in addition to the inclusion of "...profiles that aim to facilitate the visual representations for regulatory sub-processes of learning...an action plan...materials and resources to use and allocation of time...the actual use of learning resources can be indicative of learning outcomes..." Sedrakayan, et al (2019). Duin states that the recent advancements in academic and learning analytics will significantly enhance actionable intelligence for administrators and instructors, and most importantly, increase efficiency, effectiveness, and student success (Duin and Than, 2020). He goes on to further mention that the student performance dashboard views only share data related to the "what" of various student learning, and not the "why" behaviors and ultimately misrepresent true learner engagement within the class, and lacks in showing both the learner and educator if any cognitive, social, or similar achievements were made (Duin and Than, 2020). Vaitsis below depcis an overview of how we can analyze and represent certain curriculum data visually. 

Vaitsis et al. (2014), Visual analytics in healthcare education: exploring novel ways to analyze and represent big data in undergraduate medical education

There is a lack of  recent research on the usage of visual learning analytic dashboards used in the K-12 environment. Many studies have focused on the higher education sector, and I only came across 2 relevant literature reviews that focused on early learning. Confrey et al. (2018) describes a new framework, agile curriculum, "...as a means to support the ongoing revision and adaptation of teachers' curricular practices based on providing immediate data about what one's students are learning." (Confrey, 2018). In her study, Confrey used Math-Mapper, a digital learning system to further discuss and explain her new concept of an agile curriculum and the need to always revision the curriculum based on data, and the importance of using just in time feedback to help plan the next steps. Below is an example student report view from Math-Mapper that was used in grade 6. Additional literature on primary school reporting systems and data would furhter helps us understand if primary and high school teachers understand the benefit of visual learning analytics, and if students would deem them useful and beneficial as well. Interviewing K-12 parents in addition would provide further insight in design improvements that are necessary to deeply understand a younger learners' learning analytics. 

Confrey, J. (2018). The concept of an agile curriculum as applied to a middle school mathematics digital learning system (DLS). International Journal of Educational Research, 15.

Confrey, J. (2018). The concept of an agile curriculum as applied to a middle school mathematics digital learning system (DLS). International Journal of Educational Research, 15.

(Bodily & Verbert, 2017)

As the above table shows, there are extremely few research articles that focus ultimately on the end user, the student, and how they are benefiting. Visualizations can stimulate behavior in learning analytics dashboards, and the learner is informed with their progress, against any learner outcomes (Sedrakyan, et al 2019). Students usage across all educational settings must be researched, and their input and product feedback should be taken into consideration. Creating products and or features without consulting one of the most important users, and no taking into account their user experience, seems futile. As Van Haren and Harroun discussed in their research on CG Scholar's Analytics, visualizations that are clear, for each individual student and the entire class that depicts the steps to content mastery would support teachers. (Van Haren & Harroun, 2019). 

As learning analytics and educational data mining are both relatively new areas of research in the educational field, it looks like a very exciting time to be involved in the area. The latest technological advancements, and the opportunites posed for students globally, regardless of their location and or income level is a great achievement for all. Through my recent research the last few weeks in the area of designing visual learning analytics, it is clear that there are ample opportunities for research, but very few interested parties willing to learn about effectively designing dashboards for learners, whilst also involving the learner in the entire process of creating such a novel product feature that offers an astounding benefit to educators and learners alike. I would like work with education policy experts across various school levels, and help implement policies where educators and learners work together to discuss product design relating to visual learning analytic dashboards. Both parties involved have different needs to best undertand learning analytics, and if both are involved in the design process, everyone can benefit from quickly acting on data driven decision making to improve learner and educator outcomes.

To conclude, in the video below, Angela Lee Duckworth notes how grit helps us achieve our personal successes, and designing research based visual analytics dashboards for students and teachers can only further aid in personal succes.