Monday, 16 September
- Hybrid Learning Spaces - Design, Data, Didactics
- CROSSMMLA in practice: Collecting, annotating and analyzing multimodal data across spaces
- The Fellowship of Learning (Activities)
- Designing and evaluating innovative and transformational mobile learning tasks
- Gaming elements and educational data analysis in the learning design of the fllipped classroom
- Generating Actionable Predictions on Student Success: The role of Learning Design
Tuesday, 17 September
- Chatbots and Digital Assistants in Education (EduCHAT)
- The Ethics of Artificial Intelligence to Enhance Learning: Who Cares?
- Culturally inclusive learning analytics
- sysTems of Assessments for Computational ThinKing Learning
- Generating Explainable and Actionable Insights for LA
- Flipped Classroom Design via Open Tools
mLearn workshop, VIP pass needed:
As digital technologies permeate our everyday lives, technology-enhanced learning experiences are becoming increasingly ubiquitous and fluid. In such blended learning across multiple digital and physical spaces, traditional de-contextualized, log-based learning analytics may not be enough to understand the learning process, its meaning or its outcomes. Thus, analyzing evidence from multiple data sources will become increasingly needed and commonplace, if we are to extract meaning from these increasingly fluid, increasingly complex kinds of transformative learning (cf. the conference theme on transforming learning and meaningful technologies).
This workshop continues a recently-established but already very consistent tradition of workshops on multimodal learning analytics (MMLA) and across-spaces
The goal of the workshop will be to offer hands-on experiences with CrossMMLA data gathering, processing, annotating and analyzing techniques. This will allow for novice participants, to learn about the field directly by engaging in data collection, analysis and application of CrossMMLA for learning and teaching.
A growing body of research adresses linking learning design to learning analytics (Brouns & Firssova, 2016; Emin-Martínez et al., 2014; Gaševi? et al., 2017; Persico & Pozzi, 2015; Hogaboam et al., 2016; Knight & Buckingham Shum, 2014; Lockyer et al., 2013; Monroy et al., 2014; Rienties et al., 2017; Rodríguez-Triana et al., 2015; McKenney & Mor, 2015; Ruiz-Calleja et al., 2017; and Wise, 2014, Schmitz et al., 2017) Although an increased interest in and progress made toward aligning the domains of design for learning and learning analytics there still remain some issues that need to be addressed.
Share and collect experiences on several learning design methodologies and how they relate to learning activities.
The communication behaviour of us humans has changed greatly in recent years. The ubiquitous presence of digital media, networked devices and technology has created new types of conversations, discussions and information sharing. Digital platforms are also becoming increasingly important in educational practice. However, the resulting possibilities for designing lessons and supporting learners have not yet been fully exploited. Especially in the individual support of learners, current forms of education often reach the limits of teaching resources.
The main objective of this workshop is to transform the knowledge and perspectives of different experts into a guideline for further chatbots developments.
The prediction research has been based on the data from a single past course to build and test predictive models with post-hoc approaches (e.g., cross validation). However, these approaches are not valid for real-world use since they require the true training labels which cannot be known until the target event takes place (e.g., dropouts).
The first objective of this session is to teach the how to generate actionable predictions on student engagement using in-situ learning approach. Participants will be guided to compute relevant features to build a predictive model and to train the model with in-situ learning approach in Python Scikit-Learn. Learning and practising the model building and testing approaches in the session will motivate participants to build real-world interventions with predictive machine learning models.
The second objective is to illustrate the participants how to inform the feature selection with Learning Design, which is highly discarded in the literature. Features informed by learning design of the context studied can lead to more powerful models. In the session, participants will experience how to interpret the learning design to create features more relevant in the context.
The workshop aims at compiling an overview of different designs and implementations of the FC in terms of learning environments, tools and learning activities. Emphasis is put on approaches were game-based learning and/or adaptive learning has been combined with flipped learning. Finally, the workshop intents to discuss the use of learning analytics in FCs.
The aforementioned objectives are in line with the FLIP2G project objectives. The Flip2G project aims to establish a Knowledge Alliance between higher education institutions, schools and private companies that will boost skills development and introduce novel, data-driven approaches to education and training. The consolidation of all efforts will provide a transnational set of results, as follows:
• a new pedagogical method that combines PBL and flipped classroom with game-based learning
• a simulation-based serious game that supports PBL-enhanced flipped classroom processes, adaptive pathways and educational data recording
• learning designs for higher education, schools and business that support the Flip2G paradigm
• learning analytics features that produce informative insights on learning process.
More information on the FLIP2G project at: flip2g-project.eu
The workshop theme is designing and evaluating mobile learning for learners. Learners could be located in schools, post compulsory settings, informal/non-formal contexts, such as museums or community groups. The workshop focuses on the following TEL topics: innovation, transformation, disruption, mobile design and evaluation. There is much discussion about the potential of mobile technologies to transform teaching (Joan, 2013; Kee & Samsudin, 2014; authors, 2017), however, the reality is that pedagogies and learning have not changed much even whilst mobile technologies have become ubiquitous in daily life. Incorporating mobiles into educational settings in effective ways is difficult, and expectations for innovative change rarely take into account the complexity of learning, the preferences of students and the interest and motivations of teachers (authors, 2019, Jordan, 2011). This workshop explores this nexus between feasibility and innovation for mobile technology-enhanced learning in education. It firstly considers research findings on what actually is occurring in teaching and learning of school-aged students, and then presents participants with a new, research-inspired, professional learning app to guide them through this complex landscape for their own innovative m-learning task designs. Our definition of innovative mobile pedagogies is of new pedagogies that are expressly designed to take advantage of mobile device characteristics to enable effective learning to occur in ways and contexts that could not occur without mobile devices. Innovative practices are ones that are different from accepted and conventional practices, and include the effective use of new technologies (in this case mobile technologies) to promote 21st century skills of creativity, communication, collaboration and critical thinking (authors, 2010; authors, 2018; P21, 2007). Innovation suggests “new ideas or practices that are impactful and valuable to individuals or communities” (authors, 2018, p6). For this workshop, we focus on new pedagogies that will contribute to effective learning for learners. This approach is comparable to previous measures and dimensions of innovation designed by Law, Chow and Yuen (2005) who identified six dimensions of innovation and three descriptors to measure these. Drawing on our recent research (authors, 2019, authors, in-press) that used these dimensions, we note that current examples of innovative pedagogies lie on a continuum from ones that modify existing pedagogies, sometimes called sustaining (or incremental) innovations (Christensen, Horn & Johnson, 2008; Cranmer & Lewin, 2017; authors, 2018) to ones that create new practices, unlike those used previously. The latter are likely to be disruptive in nature, causing a change in paradigms, behaviours, and goals (authors, 2018), hence the term disruptive (or radical) innovations (Christensen, et al., 2008; Cranmer & Lewin, 2017). This workshop explores examples from both ends of this ‘innovation spectrum’ and sets out to help participants understand what lessons can be learned from the existing research on innovative mobile pedagogies. In particular, we will introduce 21 innovative mobile learning principles emerging from our recent research (authors 2019). Through use of the aforementioned app, workshop participants will then be guided in designing an innovative m-learning task that is informed by these principles and suitable for their own context.
The workshop will prepare participants, chiefly academics researchers, practitioners (e.g. teachers) and TEL developers, to better understand, design and evaluate the transformational potential of mobile and pervasive technologies, and to prepare for the submission of an academic article in a special edition of BJET (due 2020) which the workshop leaders will co-edit. The special edition will focus on innovation in mobile learning and will include contributions from all phases of formal and non formal education. The specific goals of the workshop include:
- Develop a deeper understanding and awareness of how to design innovative mobile learning activities (primary, secondary and tertiary contexts, including teacher education)
- Develop skills and criteria to evaluate mobile and TEL activities
- Improve own digital pedagogical skills and understandings beyond current practices
- Participate in and join a vibrant TEL community: the DEIMP community
- Broaden understanding of innovation and transformation in TEL
- Understand and be better prepared to submit to an upcoming BJET special edition on innovation and m-learning
The spaces we teach and learn in are changing. Technology is permeating physical spaces, augmenting and enhancing learning experiences. Mobile and pervasive internet-connected technology (IoT) create interfaces between virtual spaces and real-world phenomena in which big data is collected. These dynamics give rise to a growing presence of hybridity: the blurring of boundaries between distinct contexts of learning and activity, and the unexpected interleaved experiences they engender (Ellis & Goodyear, 2016; Trentin, 2016). Hybridity is not a technical issue. As Stommel (2012) notes: “The word ‘hybrid’ has deeper resonances, suggesting not just that the place of learning is changed but that a hybrid pedagogy fundamentally rethinks our conception of place”. Cook et al. (2015) identify two dimensions of hybridity: the interleaving of formal and informal social structures in an activity system, and the combination of physical and digital tools mediating individual’s interaction with the world and society: “people connect and interact through a hybrid network of physical and technology-mediated encounters to co-construct knowledge and effectively engage in positioning practices necessary for their work”. Recognition of the potential of hybrid learning spaces in promoting significant changes in learning is growing, along with increased use of hybrid learning models (do Mejía Gallegos et al., 2017). Recent work has begun exploring the nature of hybridity from a design perspective (Köppe, Nørgård, & Pedersen, 2017).
Hybrid learning spaces open opportunities and pose challenges for designers of learning experiences. Apart from the complexity of combining multiple modalities to achieve effective synergies, activities within them generate data, which can be used to monitor learning processes, potentially feeding back into them to enable “double loop learning”: awareness and control of the process of learning and teaching itself (Blaschke, 2012). Recent years have witnessed a growing interest in the promise of educational data science (EDS). In particular, there is an emerging recognition of the valuable intersection between data and educational design (Hernández-Leo et al., 2017; Mor et al., 2015; Toetenel, & Rienties, 2016). While the tradition of EDS originated in the study of virtual learning environments, we see first advances into its use in physical environments (Cukurova et al., 2017; Prieto et al., 2018). However, although the correlation between physical space design and educational effect is well established (Tanner, 2000), Learning space research is a relatively new field of study that seeks to inform the design, evaluation and management of learning spaces (Ellis & Goodyear, 2016) and EDS has not yet ventured into this domain.
Along with the opportunities that arise from these hybrid learning spaces, there are issues that require an in-depth discussion among the community of researchers, developers, and practitioners in the field. While some of these issues are well understood, others are only beginning to be explored. An educational data science perspective accentuates and amplifies both the opportunities and the challenges.
This workshop is structured as a flipped classroom experience which introduces participants to flipped classroom design through different tools and OER (open online resources) and provides practical support to develop and apply new flipped classroom design skills.
Using open online sources is a cost-effective approach that can lead to positive learning outcomes. We will provide you with an overview of existing free online tools and work with these in the most effective way for flipped classroom development. Tools are not limited by application only in flipped classroom settings and can be used for digitizing one of the element of your class both in higher education and professional trainings
Among others these tools will include: Powtoon for animation development; Screencast-o-matic for video development; EdPuzzle, TedEd- for video with embedded questions; FlipGrid- for video homework set up; Thinkific for online course platform development; Strikingly, Wix - for website development and course sharing; as well as free content on platforms like Ted.
Participants will have an opportunity to familiarize yourself with materials before the workshop here (https://www.coursecrafting.org) and workshop will be devoted to activities which primarily refer to trials with different tools (individually and in a group-setting).
In this workshop the main goal is to familiarize participants with different options how to design their flipped classroom using the most resource effective technologies. There are variety of different technologies which can be used without any financial investment or with minimum costs The flipped classroom settings are not necessary condition. The tools can be used also for designing digital content for any other type of blended or online learning.
While Artificial Intelligence is increasingly being used to support teaching and learning, most AI in education research, development and deployment takes place in what is essentially a moral vacuum. In short, little research has been undertaken, no guidelines have been provided, no policies have been developed, and no regulations have been enacted to address the specific ethical issues raised by the use of AI to enhance learning (Holmes et al., 2019; Luckin et al., 2016).
This is of particular relevance for the EC-TEL community, which is why we are proposing a workshop for EC-TEL 2019: The Ethics of Artificial Intelligence to Enhance Learning: Who Cares?. This will be an opportunity for researchers who are exploring the ethical issues of AI being used to enhance learning to share their insights, to identify key ethical issues, to map out how to address the multiple challenges, and to inform best practice. The overarching aim will be to help establish a basis for meaningful ethical reflection necessary for innovation.
The objective of this workshop is to enable the EC-TEL community to explore the issues, and take a lead in this important debate about the ethics of AI to enhance learning.
Within the Learning Analytics (LA) community, the idea that a “one size fits all” paradigm does not lead to effective LA tool designs has become widely accepted, but there is still a big question mark over what factors that define the “right size” for every learner. During this workshop, we wish to explore how to design inclusive LA tools in order to minimise cultural barriers.
Raise awareness on the effects of culture on learning and beliefs about learning and implications on the:
- acceptance and use of LA tools
- design requirements of LA tools
- potential for reusability and the need for adaptation of LA tools across different cultural contexts
- need for adaptation of LA tools within cultural heterogeneous environments.
Computational thinking (CT) is considered as a key set of skills that must be acquired and developed by today’s generation of learners. There is now consensus that CT should be taught not only in CS classrooms but also in the context of STEM and other subjects (Grover 2017). However, an agreement is missing on CT assessment. Grover and Pea make the gravity of this gap clear: “Without attention to assessment, CT can have little hope for making its way successfully into any K-12 curriculum” (Grover and Pea 2013).
- Provide the opportunity to the TEL researchers to discuss and share their ideas on CT assessment
- Facilitate interdisciplinary collaboration among the participant researchers
- Involve teachers and learning experts, who can bring their every-day working experience into the group and contribute to the discussion
- Draft a post-workshop report that summarizes the main workshop findings
Data about learning are abundant. Algorithms for analyzing these data are as well. The growth of learning data, the development of algorithms and AI, and the combination with learning sciences gave birth to the domain of Learning Analytics (LA). LA is about `collecting traces that learners leave behind and using those traces to improve learning‘ (Eric Duval). Verbert et al. (2013) showed however that the actual improvement of learning is at the ‘impact’ level (level 4) and that this improvement does not follow directly from the awareness of the data (level 1), but first has to go the levels of self-reflection (level 2) and sense-making (level 3). To generate information out of data, machine learning and AI have provided many algorithms that look for patterns in the data. Visualization techniques and Learning Dashboard can provide a visual entry to explore patterns in data or to explain the outcome of these algorithms.
Currently, the LA domain is maturing and often provides insights in and recommendations regarding learning and teaching behaviors. These insights by itself are not enough. Before the insights can actually impact learning they have to be interpretable and actionable. To be interpretable, the outcomes of the data analysis, visualization, and/or algorithms have to be tailored to the end-users. While advanced visualization and/or machine learning techniques might create accurate and trustworthy insights and recommendations, they will not be per se trusted by the user. Opening the black-box of learning analytics to the user, in a user-tailored fashion is the first step towards obtaining interpretable insights and explainable recommendations. Approaches for obtaining transparency, trustworthiness, persuasiveness, and effectiveness are key.
An promising approach is to add interactivity and even configurability to LA dashboards and models. In other words, the actions of the users and their own insights affect the data sources being used, the visualizations presented and even the workings of the analytic models.
The next step is to translate insights into `actions’. Many insights obtained from learning analytics are however not directly actionable for the user. While learning that male students and students from a more socially-challenging background are more likely to fail can be useful for policymakers, it will not be for an individual student. He/she can’t translate such an insight to an individual learning action. When actionable insights and recommendations can be created for learning analytics data and/or algorithms that are tailored to the end-users, they will have the potential to create actual impact. Actionability works in both ways and can greatly be improved by integrating the LA into the pedagogy of study programs and individual courses. By making instructors aware of how LA can interact with their courses, they can contribute to better the recommendations of resources to students.
The evolution towards actionable insights and explainable recommendations is urgent, as recent data protection and privacy regulations (EU GDPR and CCPA (California Consumer Privacy Act)) stipulate that transparency is a fundamental right. Even more, they state that each user has the right to withdraw him/herself from automatic decision making and profiling. So, it is now time to act and to jointly work towards transparent actionable insights and explainable recommendations in the domain of Learning Analytics!
The motivation of the workshop is to advance the research and practices around the creation of actionable insights and explainable recommendations in the domain of Learning Analytics.
Note: you can only participate in this workshop with a VIP pass for both conferences
This is the first international workshop emerging from the Safety in Smart Cities research programme. The emerging concept of smart cities with digital information systems, sensor-based infrastructures, autonomous vehicles, robotic appliances, internet of things appliances affects many dimensions of the life of the city’s inhabitants. One of these dimensions is safety. While the infrastructures emerging within smart cities also contribute to safety, they may also impose new risks, which require adapted and changed behaviour of citizens. Many aspects of smart city environments are also not communicated transparently enough, so that misleading, conflicting, and wrong information may be distributed leading to fears, irrational behaviour and unnecessary risks.
This discrepancy between perceived and actual safety risks, the availability of new situations confronting citizens of smart cities, the required behavioural changes of citizens acting in smart city environments, and the availability of new data sources emerging from smart cities ask for new approaches of TEL addressing these aspects.
The proposed workshop aims to discuss, which TEL approaches may be needed to educate inhabitants of smart cities about benefits and risks of smart cities, behavioural impacts of smart cities, and safety-related aspects. Which elements of the smart city can contribute to safety-related education? Which societal effects can we expect from safety-aware smart cities?