Learning Analytics: from A to ... A

The document discusses paper prototyping and its benefits for getting early feedback on designs. Paper prototyping allows designers to create low-fidelity prototypes using paper, scripts, or other basic materials. It discusses different types of prototyping like paper, visual programming, and web prototypes. The document provides several examples of "before and after" redesigns and emphasizes the importance of understanding users and goals when prototyping interfaces.

The document discusses the "Snowflake Effect" in learning, which is the idea that personalized learning experiences can be provided for every person daily, tailored to their specific needs, interests, context and ways of learning. It promotes "mass personalization" or "meLearning" where billions of individualized "markets of one" can be supported through open learning infrastructures and interconnectivity on a global scale.

The document discusses the concept of abundance and the "snowflake effect". It notes that if a piece of paper was folded 50 times, it would be thicker than the distance to the sun, demonstrating exponential growth. It also references the abundance of information available online, such as the billions of images and videos now accessible. The document concludes by stating that the best way to predict the future is to invent it.

This document discusses learning dashboards and learning analytics. It introduces the Learning Analytics Reflection and Awareness environment (LARAe) dashboard, which visualizes learner data and badges to improve awareness and reflection. Several versions and features of the LARAe dashboard are presented, including the use of badges, videos demonstrating its capabilities, and its integration with emotion detection tools. The dashboard is aimed to help learners monitor their own learning through visualized traces and badges.

The document discusses learning analytics, which involves collecting data about learners' online activities and using that data to improve learning. Specifically, it tracks how tracking personal data through technologies like fitness trackers can inspire self-reflection and awareness. However, there are issues with determining what types of data best reflect learning versus just measuring time or outputs. The presentation advocates discussing with learners what data is being collected, why, and how it will be used.

1) The document discusses learning analytics, which involves collecting student data from online learning environments to improve learning outcomes. 2) Key aspects of learning analytics discussed include aims like improving motivation, audiences like learners and teachers, and architectures to scale analytics across different systems. 3) The author advocates an open learning analytics architecture using existing web standards to enable large-scale and cross-platform analysis of student data with the goal of better understanding learning processes.

Erik Duval presented on the Quantified Self movement at an investors meeting in Leuven, Belgium on June 4, 2013. The presentation discussed how individuals are increasingly tracking various metrics about their lives through technology, referred to as "digital exhaust", and how this data could be used for learning analytics, health monitoring, and driving innovation through ubiquitous tracking opportunities. Only the beginning of these types of self-tracking applications was noted.

Erik Duval gave a presentation on open learning analytics at MBO City in Ede, Netherlands on November 25, 2013. Open learning analytics involves collecting data traces that learners leave behind as they learn and using that data to improve the learning process. As learning moves online, more data is available which can provide insights if analyzed using techniques from data mining, visual analytics, and by combining algorithms with human judgment. The goal is to leverage analytics to provide awareness and feedback to learners and teachers to support self-reflection and improve communication.

This document discusses multimodal learning analytics and the challenges associated with it. It presents a case study called the PELARS project, which collects and analyzes data from various modalities like computer vision, mobile devices, and workstations to understand hands-on STEM learning. The challenges discussed include getting meaningful data from real classrooms, analyzing messy and incomplete multimodal data, and developing visualizations that provide useful insights. Real-world applicability and scalability of multimodal learning analytics approaches are difficult but important open questions.

This document discusses learning dashboards and analytics. It introduces learning analytics as collecting learner traces to improve learning. Dashboards focus on providing awareness to learners about their own activities and progress. They aim to help learners reflect and make sense of their data, which can lead to behavior change. Effective dashboards consider what types of data to present, such as time spent, resource use, communication, and artifacts. They also consider how to visualize data for perceived usefulness and findings. Issues include data architecture, interoperability and privacy.

Dr. Hendrik Drachsler is an associate professor researching learning analytics, personalization, recommender systems, and mobile learning. His research focuses on applying these topics in schools, higher education, and medical education. He discusses learning analytics frameworks and models, challenges around educational data standards and privacy, and the importance of developing learner competencies like data literacy and agency.

1) The document discusses learning analytics, which is about collecting data traces that learners leave behind from online activities and using that data to improve learning. 2) As more learning moves online, more data traces can be collected from students' activities over time, not just at the end of a course. This data can provide insights into the learning process. 3) Learning analytics focuses on using data to provide awareness to learners and teachers about the learning process and empowering students through reflection and sense-making of their own data.

The document discusses learning analytics and focuses on collecting and analyzing student data traces to improve learning. It references empowering students through self-reflection and awareness of their own learning patterns. Key issues discussed include student ownership of data, capturing meaningful traces, and evaluating the impact of learning analytics approaches. The document promotes the upcoming LAK13 conference on learning analytics and knowledge to be held in Leuven, Belgium in April 2013.

This document discusses the concept of open education and learning analytics. It promotes teaching students to solve unknown problems using unknown technologies through open courses and authentic problems. It advocates for continuous monitoring and open analytics dashboards to provide awareness and self-reflection for students. Finally, it addresses issues of open accreditation, privacy, and the potential for more ubiquitous learning both in physical and online spaces using wearable technologies.

This presentation was used Tinne De Laet, KU Leuven, for a keynote presentation during the event: http://www.educationandlearning.nl/agenda/2017-10-13-cel-innovation-room-10-learning-and-academic-analytics organised by Leiden University, Erasmus University Rotterdam, and Delft University of Technology. The presentations presents the results of two case studies from the Erasmus+ project ABLE and STELA, and provides 9 recommendations regarding learning analytics.

This document discusses learning analytics in mobile learning. It notes that learning analytics involves collecting learner data traces from activities and using them to improve learning. Mobile devices now provide opportunities for always-on data collection through sensors. However, issues around privacy, transparency and the meaning of the data must be addressed. The presentation explores questions around what types of data learning analytics in mobile contexts could capture and how that data could provide insights into learning itself, not just activities.

Maandag 9 november Plenair Titel: De zes dimensies van learning analytics Spreker(s): Hendrik Drachsler Zaal: Rotterdam Hall

This document provides an overview of a workshop on using learning analytics to improve student transition and support in the first year. The workshop was delivered by the ABLE and STELA projects in partnership. It begins with introductions of the presenters and a discussion of the workshop structure. Next, the document explores definitions and concepts of learning analytics through short discussions and examples. It then highlights examples of learning analytics projects and implementations at partner institutions like Nottingham Trent University, Leiden University, and Delft University of Technology. The workshop also included an exploration activity where participants discussed goals and interventions for a hypothetical learning analytics project. Finally, the document outlines three case studies that workshop groups worked on, with an emphasis on presenting results

Erik Duval gave a presentation on learning analytics for visualization and recommendation. He discussed how learning analytics can be used to track learner attention and display that data through visualizations. This allows learners and teachers to gain insight into the learning experience. Duval presented examples of dashboards and tools that visualize learner activities and link them to goals to promote self-reflection. He also discussed how recommendation systems can be improved using learning analytics data and federated search. However, Duval noted there are also dangers if too much data is collected without transparency and if users become trapped in "walled gardens" controlled by companies.

Hendrik Drachsler presents on envisioning the future of learning analytics at an education conference in the Netherlands. He discusses the LACE project, which aims to integrate communities working on learning analytics in schools, workplaces and universities. The presentation outlines four visions for the future of learning analytics in 2025, including analytics being essential tools for educational management, supporting self-directed autonomous learning, rarely being used due to data privacy issues, and personalizing education through adaptive recommendations. Drachsler conducts a workshop to gather feedback on these visions from conference attendees.

This document discusses Universal Design for Learning (UDL), which aims to make learning accessible to all students by providing multiple means of representation, engagement, and action/expression. It notes that UDL is grounded in neuroscience research showing people learn in different ways. The three principles of UDL are outlined, and examples are given of how they can be applied in the classroom using technology and other tools. Implementing UDL is argued to benefit students by better addressing their diverse needs and strengths.

Keynote presentation at Edmedia 2018 conference: https://www.aace.org/conf/edmedia/speakers/. Results of Erasmus+ projects ABLE (www.ableproject.eu) and STELA (www.stela-project.eu) on learning dashboards for supporting first-year students.

This document summarizes a presentation on learning analytics given at the ALT-C 2014 conference in Warwick, UK. It discusses the Learning Analytics Community Exchange (LACE) project, which is a 24 month EU support project with 9 partners focusing on implementing learning analytics in schools, higher education, and industry. The presentation defines learning analytics as the measurement, collection, analysis and reporting of learner data to understand and optimize learning. It also discusses cultural and technical challenges around topics such as data privacy, change management, and making insights actionable. Examples of learning analytics tools described include SNAPP for social networks and LOCO-Analyst.

The document outlines the agenda for an eCloud workshop taking place on May 13th, 2015 in Amsterdam. It includes the following sections: - A welcome and logistics session from 09:30-09:40 - Participant introductions from 09:45-10:30 where participants will briefly introduce themselves - Breakout group sessions from 11:30-13:00 where participants will discuss personas, scenarios, tool identification and evaluation sessions - Presentation and discussion of the breakout group results in the afternoon sessions.

For my course on information visualisation...

This document contains the notes from a lecture on information visualization given by Erik Duval. It discusses state-of-the-art evaluation methods for information visualization, including controlled experiments, usability evaluations, case studies and natural environment studies. It also mentions challenges with evaluation and replicating results. The notes raise questions about the students' work and plans for a final demo the following week.

For my course on information visualisation...

The document discusses different methods for evaluating information visualization, including controlled experiments comparing design elements, usability evaluations, and case studies analyzing real tasks in a natural environment. It notes that redesigning visualizations can also be a form of evaluation, and references several papers on visualization evaluation and redesign techniques.

Presented on 27 April 2015

This document discusses the relationship between social media and science. It explores how scientists are using social media platforms like ResearchGate, Academia.edu, Mendeley and Twitter to collaborate, share research, and increase the visibility of their work. While social media provides opportunities for exposure, discovery and crowdsourcing, it also risks fragmentation of attention and blurring the boundaries between work and personal life. The document concludes that science has always been social, and social media both enhances and complicates its social nature.

The document discusses Erik Duval's presentation on Edward Tufte's principles of data ink design. It outlines Tufte's key principles: showing the data above all else, maximizing the data-ink ratio by removing non-data ink, erasing redundant data ink, and revising and editing visualizations. The data-ink ratio refers to the proportion of ink devoted to displaying non-redundant data information. The principles aim to clearly display the maximum amount of data with the minimum amount of graphical elements.

For my infovis students...