learning metadata
LAK 2013: Third Conference on Learning Analytics and Knowledge
This third conference on learning analytics will be designed to bring the many voices involved in leveraging the availability of data about learning with powerful computational, representational and visualization techniques into dialogue in a “middle space” under the overarching theme of “Dialectics in Learning Analytics”.
The first two conferences have established the range of issues and approaches of concern in leveraging the availability of data about learning with powerful computational, representational and visualization techniques. This third conference will be designed to consolidate the field by bringing these many voices into dialogue in a “middle space” under the overarching theme of “Dialectics in Learning Analytics,” which has these facets:
The Middle Space: The conference will explore the “middle space” within which Learning and Analytics intersect, and seeks proposals for papers and events that explicitly connect analytic tools to theoretical and practical aspects of understanding and managing learning.
Productive Multivocality: Learning analytics is multidisciplinary, drawing on theories and methods from diverse research traditions. Our community includes educators, learning scientists, computer scientists, administrators, and policy makers, among others. The middle space serves as a topical “boundary object”, enabling productive discourse between these many voices.
The Old and the New: We are facing a centuries old problem: to improve learning, but we are trying to solve it using a new set of tools, not available before. We address these problems in the city of Leuven: centuries old, lively new.
Spreading the Word about Linked Data in Education
LinkedUp aims to push forward the exploitation of the vast amounts of public, open data available on the Web, in particular by educational institutions and organizations. LinkedUp will organise the LinkedUp Challenge to realise personalised university degree-level education of global impact based on open Web data and information. Drawing on the diversity of Web information relevant to education, this aim requires overcoming substantial challenges related to Web-scale data and information management involving Big Data to offer personalised and accessible education services.
Challenge Initialisation and Development (WP1)
LinkedUp will launch an open challenge to identify and promote innovative uses of open Web data in educational contexts. WP1 is responsible for the design and timely execution of the challenge and the interaction with other LinkedUp activities (such as data curation and evaluation). To this end, WP1 will define the main challenge tracks, criteria and incentives as well as the dissemination strategy (jointly with WP4). The LinkedUp Challenge will be organised into different tracks (from an Open Data Challenge up to more specific task-oriented challenges) which will run through different stages, where access will be possible to new participants at the start of each stage.
Key outcomes:
- Web data success stories: the LinkedUp Challenge will identify and promote highly innovative applications and technologies which exploit open Web data in ways which significantly expand the current state of the art. The latter includes technical dimensions such as scalability or performance as well as non-technical aspects related to legal, privacy or usability issues.
- Open challenge framework: a reusable competition framework which will be established as periodic series of competitions (defined in terms of timelines, categories, requirements, stages, tracks and incentive structure)
- Best practices & lessons learned: throughout the project, WP1 will refine its approach and capture a set of best practices which will emerge throughout the competition
Evaluation Framework (WP2)
One of the main outcomes of the LinkedUp project will be an Evaluation Framework (EF) that can be reused and instantiated to evaluate Open Web Data applications in particular in the educational domain. WP2 will develop the EF that consists of predefined evaluation criteria, metrics, methods and benchmarks for the assessment of open data-based technologies and data itself. Evaluation dimensions include technical ones (such as performance, scalability, precision) as well as non-technical ones (qualitative criteria, legal and licensing issues, usability criteria). The EF will be developed with the help of the Group Concept Mapping (GCM) method and the support of an inter-disciplinary panel of experts. Within the GCM method, experts have to agree on a collection of specific evaluation criteria and their indicators.
Key outcomes:
- Evaluation framework (EF): a set of evaluation criteria, metrics, methods and benchmarks for the evaluation of data-driven applications and data. The EF will be public and available for any party to use and expand.
- Specific EF instantiations for the LinkedUp challenge tracks: specific subsets of the EF will be developed for the particular assessment in each track (open track vs focused task track). Hence, guidelines for qualitative assessment by experts (open track) will be provided as well as a set of automated assessment metrics and methods (focused task track).
- Evaluation results: in addition, the outcomes of LinkedUp evaluation activities will be published and made available. This will result in a set of publicly available data and technology quality assessments and resulting benchmarks.
Deployment Support: Data, Guidance, Infrastructure (WP3)
LinkedUp relies on a base infrastructure, both technical and organisational, to support the development of educational applications exploiting Web data. While a wide variety of educationally relevant data exists on the Web, most prominently,LinkedUp data curation activities will produce a data catalogue and repository which will offer access mechanisms to a wide range of datasets of relevance to educational scenarios. WP3 will also provide development support for external developers and will collaborate with them on tackling non-technological issues associated with deploying web data applications in an educational environment (for instance, legal and organisational aspects).
Key outcomes:
- LinkedUp data catalogue & registry: a public catalogue of categorised and described educationally relevant datasets
- LinkedUp data infrastructure: public access mechanisms to educationally relevant datasets (endpoints & APIs)
- LinkedUp support environment: supporting developers in exploiting educational datasets
Community building and dissemination (WP4)
LinkedUp will catalyse an active, diverse and well-connected community in the area of open linked data for education, including open Web data and resource evangelists. We will bridge the research and business communities, ensuring that innovative results and knowledge from academia are transferred to practical applications, eg. in a business context. We will enable and encourage content and data providers to contribute new material to LinkedUp through events, tools, and documentation. In order to facilitate learning about linked data for education, we will create a Handbook on Open Data in Education, a resource for educators, web data providers, and adopters.
Key outcomes:
- A network of practitioners and experts on open data in education
- Events and workshops to encourage understanding and uptake of open data in education
- Handbook on Open Data in Education: We will collaboratively develop a handbook, gathering best practices on how to use open data to meet educational needs. This will include use cases, tips and tricks for finding data and tools, and guidelines on using data and tools.
Exploitation, Exit and Sustainability (WP5)
The goal of this work package is to develop large-scale scenarios and use cases for the deployment, evaluation and exploitation of the Web data-based application. Use cases serve two main purposes: providing scenarios for the actual deployment of LinkedUp Challenge submissions and to prepare and implement an exit & sustainability strategy for the long-term exploitation of the project results. The latter in particular ensures the persistence and long-term availability of the competition and evaluation framework produced in LinkedUp.
Key outcomes:
- LinkedUp Large-scale use case scenarios: real-world scenarios for the deployment of the applications developed during the LinkedUp challenge, covering public as well as educational sector
- LinkedUp Exit and sustainability plan: to ensure the persistence and long-term availability of the competition and evaluation framework produced in LinkedUp
- LinkedUp Show cases: will show the application of LinkedUp technologies to use case scenarios, featururing a meaningful subset (software, data, etc.) of the functionality characterising the project demonstrator(s)
The Geoscience Concept Inventory WebCenter provides new means for student assessment
As an instructional tool, the WebCenter contains concept inventory questions that are carefully designed to ascertain student’s conceptual understanding in a range of geology subtopics. The WebCenter’s customized LON-CAPA platform facilitates the inclusion of digital images created by ICT technologies to assess student learning. The WebCenter’s online venue encourages community participation in assessment development by allowing users to review existing questions and submit their own. Furthermore, the WebCenter’s testing function provides an authentic online assessment experience that aligns with ICT practice and takes advantage of its technological capabilities to provide immediate feedback and detect fine-grained data such as time-on-task.
Currently, user activity in the portal is limited to viewing and student evaluation on a small scale, with only a small fraction participating in the development of new concept inventory questions. Thus, it may be that on-site teacher training workshops are needed to help initiate collaborations and use of the technology. However, the WebCenter has already made an impact with its online, open-source nature; encouraging participation from around the globe, as evidenced by the number of users (n=130) and range of institutions using the GCI. Statistics collected via online testing with a variety of student populations