Are #xAPI Learning Designers Collecting Useful Data?
At the ending of 2013, IBM revealed its predictions for five big innovations that will change our lives within five years. The number one on the list is “The classroom will learn you“.
Meyerson said that this year’s ideas are based on the fact that everything will learn. Machines will learn about us, reason, and engage in a much more natural and personalized way. IBM can already figure out your personality by deciphering 200 of your tweets, and its capability to read your wishes will only get better. The innovations are being enabled by cloud computing, big data analytics, and adaptive learning technologies.
This claim states that Learning Analytics(LA) will change our learning. We agree. But…
Actually the learning happens everywhere. Open learning settings, like Massive Open Online Courses(MOOCs), independent studies, social learning, hands-on experiences and collaborations, which occur in decentralized, distributed teaching and learning networks, all bring challenges to LA. So, how do we track all learning experiences?
#xAPI is a way to use human-readable data streams to track experiences by using an actor, verb, and activity structure that mimics lang! — Andy Johnson
#xAPI is a specification defining an interoperable data structure and transport mechanism for learning records across platforms. — Nikolaus Hruska
#xAPI is a technology that makes it possible to track meaningful (and granular) actions in the context of learning & performance. — Aaron Silver
#xAPI is the future of learner analytics. Being device- platform-agnostic, providing tangible data, from any type of learning activity. — Sreekanth C
What’s tricky about that “all learning experiences are trackable“? It sounds wonderful until you really put your hands on designing some xAPI statements, you’ll just find it’s deeper than you thought.
Too many data will make finding useful data like finding a needle in a pile of hay. And, generating tons of statements like “Actor”+”experience“+”Object” offer very limited insights. Each statement is like a tiny pixel on the whole picture, if all pixels can’t be properly related and organized, we can’t make sense out of them.
So we need to have the final picture in mind, and think backward. What are the objectives and learning goals? How do we want to build the reporting? How can xAPI data be used to approach our purposes through iterations?
What we really want to capture are the learning moments, contexts and quality! And the learning activities are distributed across multiple sites, venues and multiple identities.
A picture of Revised Bloom’s Taxonomy could start up the thinking. A correct verb is the first step in every statement. Also be conscious of exisiting verb repository: ADL vocabularies, Tin Can API registry.
Although xAPI uses human-readable display language, learning designers need to understand the structure and its “grammar”. Not to say the learning design part is an art. How about starting from writing statements about your own learning moments? (not necessarily recording immediately when the events are happening, it could be a reflection) With the purposes of building meanings out of all tracking data in mind, you should create a meaningful picture at the end. That’s your artwork.
Our mini course “Learning Architect” is actually inviting you to co-design the learning experience and the reporting capability. Your data are in your own account. Please tweet your questions and thoughts with #xAPI, we’ll follow up and hopefully pull them together with your learning records in the LRS.
Four dimensions(4C) of learning will be depicted leveraging xAPI as the witnessing mechanism, stayed tuned.