The program is as follows:

• 11:15 – 11:30: Registration
• 11:30 – 12:50: Poster and demo session & lunch
• 12:50 – 13:00: Welcome by Katrien Verbert
  
• 13:00 – 13:30: Closing the Loop in Education – Carlos Delgado Kloos
  

  In control systems there is a feedback loop that compares the obtained output with the expected one. This allows to correct deviations. Robust theories exist and there are many systems that incorporate them in their operation. To mention just a few of them, consider thermostats, power steering systems, SCADAs, and many others.
  Learning frameworks have been in essence open-loop systems or systems with a very week and broad feedback loop. At the end of a course, if the learning of a student reaches a minimum, the student can advance to the next course. If not, the course has to be repeated. There is the feedback loop but at a quite broad level. However, one can include much faster feedback loops into the learning design in order to achieve better learning results. Examples of good learning frameworks with fine-grain closed loops are formative assessment, learning analytics based interventions, mastery learning, improvement of teaching material through effectiveness analysis, etc.

  In the talk, we will analyse educational systems from this point of view and give appropriate examples

• 13:30 – 14:00: Learning from Learning Analytics: risky assumptions about the efficacy of feedback for learners – Ed Foster

  If assessment is the ‘engine that drives learning, feedback is the oil that lubricates the cogs of understanding’ (Race, 2006). Arguably, learning analytics provides learners, lecturers and study advisers with new levels of sophisticated feedback data in a far timelier manner than has historically been available. This ought to be transformative. Ought to be.

  Nottingham Trent University implemented a whole-institution learning analytics resource in 2014/15. The Student Dashboard provides information to both students and staff about how students are engaging with their studies. Our experience is that around 40% of students frequently engage with the resource, have the highest engagement and are most likely to progress/ achieve the best grades. It appears that for students with sophisticated approaches to learning, learning analytics is an absolute boon, however far more work is needed to develop students’ capacity to learn from feedback presented directly by learning analytics, or mediated by staff. Assuming that all students can learn from learning analytics feedback is at the very least a risky assumption.

  (RACE, P., 2006. The Lecturers Toolkit: A practical guide to assessment, learning and teaching. Oxon: Routledge.)

• 14:00 – 14:30: Feedback automation for complex learning tasks – Monique Snoeck
  Teaching complex subjects to students generates a high demand for personal, elaborative and cognitive feedback.  In the presence of large number of students this quickly becomes a too time consuming task for a teacher. In this presentation I will explain two approaches to feedback automation for teaching conceptual modelling. The first approach constitutes of feedback features built into a conceptual modelling tool. The second approach makes use of standard features in the edX platform.  Both approaches rely on the systematic analysis of student errors. These errors are then used to create cognitive feedback and exercises for part-task practice with corresponding feedback. The feedback has been shown to have a positive impact on learning.
• 14:30 – 15:00: Scaling personalized feedback – Yi-Shan Tsai
  

  Feedback is a crucial part of communication between students and teachers in terms of clarifying expectations, monitoring the current progress of learners, and moving towards desired learning goals. However, there is substantial evidence showing that higher education struggles to deliver consistent, timely, and constructive feedback to meet the needs and expectations of students. The inadequacy in delivering effective feedback to students is partly due to conflicts between an increasing focus on ‘massiveness’, ‘inclusiveness’, and ‘personalisation’ in higher education and yet unmatched capacity of staff to produce feedback that speaks to the needs of individual students. OnTask is a semi-automated feedback tool, which uses ‘if…then’ rules to help teachers compose personalised feedback for a large cohort of students based on parameters relevant to the course design. In this talk, I will talk about key elements of effective feedback and how we can leverage feedback practice using OnTask.

• 15:00 – 15:30: Break and poster and demo session
• 15:30 – 16:00: Learning analytics and feedback: not only for students – Jan Elen
  Learning happens in an environment. Learning results from the interaction between characteristics of the learners and characteristics of the environment. In learning analytics research the focus is on the learner. Using data on learning behavior learning analytics provides feedback to the learner in view of predicting learning outcomes and/or of encouraging learners to change their behavior. The characteristics of the learning environment largely remain out of scope.
  In this contribution two other potential uses of learning analytics data are suggested. A first proposal pertains to using learning analytics data to provide feedback on the validity of instructional design models. The second proposal inverses the current learning analytics logic by using learning analytics data to provide feedback on the quality of learning environments rather than of students.
• 16:00 – 16:30: Interactive recommender systems and dashboards for learning – Katrien Verbert
  Researchers have become more aware of the fact that effectiveness of recommender systems goes beyond recommendation accuracy. Thus, research on these human factors has gained increased interest, for instance by combining interactive visualization techniques with recommendation techniques to support transparency and controllability of the recommendation process. In this talk, I will present our work on interactive visualizations to enable end-users to interact with recommender systems. The objective is two-fold: 1) to explain the rationale of recommendations as a basis to increase user trust and acceptance of recommendations, and 2) to incorporate user feedback and input into the recommendation process and to help them steer this process. In addition, I will present the results of several user studies that investigate how such explanations and user control interact with different personal characteristics, such as expertise and visual working memory. I will also present concrete applications in the field of learning analytics and job recommender systems that use visualization techniques to support interaction with recommender systems.
• 16:30 – 17:00: Bridging the gap between educational research and practice with learning analytics: insights from studies in primary and secondary education – Frederik Cornillie
  In educational research, the collection and analysis of data on technology-mediated learning behaviour have long been used with a view to answering fundamental questions on learning. Typically, however, these data were collected in rather sterile learning tasks and less realistic contexts, compromising the ecological validity of the findings. Mobile technologies and learning analytics have now made it possible to study fundamental questions on learning in a variety of more realistic and less controlled environments. The results of such research can inform educational technology practice.  In this talk, we present two empirical studies informed by learning learning analytics that were carried out in primary and secondary education classrooms. The first study focuses on automatization processes and the role of corrective feedback in language learning; the second study focuses on task complexity and adaptive learning in maths practice. The findings provide instructional designers and teachers with insights on how to enhance and personalize the learning environments for their pupils.
• 17:00 – 17:30: Learning Dashboards for feedback at scale – Tinne De Laet
  Learning analytics is hot. But are learning dashboards scalable and sustainable solutions for providing actionable feedback to students? Can learning dashboard be applied for feedback at scale? Is learning analytics applicable in more traditional higher education settings? This talk will share experiences and lessons learned from three European projects (STELA, ABLE,  and LALA ) that focuses on
  scalable applications of learning dashboards and their integration within actual educational practices. Can learning dashboards deployed at scale, create new learning traces? The talk will challenge your beliefs regarding “chances of success”, predictive models, and explainable interpretations.

• 17:30 – 17:40: Closing