Data format and interoperability
Data interoperability for LA is important. So important that there is an ISO standard devoted to this very topic (https://www.iso.org/standard/68976.html) and an ongoing working group ().
The Data Interoperability Standards Consortium (DISC) is working towards achieving true syntactic and semantic data interoperability for lifelong learning. You can find out more about their work here: http://datainteroperability.org/about-disc/
Currently there are two modern specifications for storing Educational Data. They are both very similar, being based on the Activity Streams specification (see e.g. http://tincanapi.com/tin-can-vs-activity-streams/). This folder will help you to understand what is going on in this important area right now and how it affects LA.
Experience API (xAPI) v.1 was released by Advanced Distributed Learning in 2013 under an Apache license and is completely open source (i.e. with anyone able to contribute to the specification if they choose to get involved). You can find out about it on:
- The ADL webpages: https://www.adlnet.gov/adl-research/performance-tracking-analysis/experience-api/xapi-technical-specifications/
- GitHub where you can find the full (and most recent) specification https://github.com/adlnet/xAPI-Spec/ (A snapshot of the specification is attached below, but will become outdated quickly.)
More recent work on xAPI has started looking at profiles and how they might be used to ensure semantic interoperability of xAPI data. For example, the Companion Specification for xAPI (https://adl.gitbooks.io/companion-specification-for-xapi-vocabularies/content/) discusses the way in which the semantic web can be used to facilitate interoperable vocabularies and semantic structures.
IMS Caliper was released in 2015 under a ??? license, however, you need to be a member of the IMS consortium to contribute to the specification, and so there are less opportunties to influence. However, all major LMS vendors and other large tool providers tend to be contributing members of IMS and so it is an important spec to keep an eye on if you are aiming for data interoperability. You can find out more about it at this set of pages: https://www.imsglobal.org/activity/caliperram
Luckily, work is currently underway to ensure that these two specifications play nicely together (yay!) Hopefully this will mean that we do in fact achieve the lifelong educational data interoperability. More information about this set of initiatives can be found at these pages:
Interoperability and LA
This folder contains the following papers about these specifications and how they have been used in LA:
- Bakharia, A., Kitto, K., Pardo, A., Gašević, D., & Dawson, S. (2016, April). Recipe for success: lessons learnt from using xAPI within the connected learning analytics toolkit. In Proceedings of the Sixth International Conference on Learning Analytics & Knowledge (pp. 378-382). ACM.
- Griffiths, D., Hoel, T., Cooper, A., (2016) Learning Analytics Interoperability: Requirements, Specifications and Adoption, LACE Deliverable D7.4. Available at: http://www.laceproject.eu/wp-content/uploads/2016/01/LACE_D7-4.pdf
You can also go and find out about large ongoing projects that attempt to unify data from many different sources at these pages:
- Apereo has an open learning analytics initiative that you can find out about here: https://www.apereo.org/communities/learning-analytics-initiative
- Apereo also recently hosted a webinar on Learning Analytics interoperability which is available on youtube here: (The video is available in the files below.)
- Jisc's work effective learning analytics project has an ongoing blog: https://analytics.jiscinvolve.org/wp/ but you can also find out a bit about the technical details of their platform here https://analytics.jiscinvolve.org/wp/2016/06/28/a-technical-look-into-learning-analytics-data-and-visualisations/