Why Do We Need Interoperability Anyway?

By Kirsty Kitto, Senior Lecturer in Data Science, University of Technology Sydney

Many educational institutions and workplace training units have invested a large amount of money in purchasing various educational technology solutions. Often this is only the start of their investment. They also need to work on integrating new acquisitions with other information systems, and ensuring that legislated reporting requirements and professional obligations are met. Often these systems end up very tightly integrated, it is such a nightmare to replace one item that change does not come without a lot of pain. So, it is easy to see that embracing new EdTech solutions comes at a cost… is interoperability important enough to go through the pain that might be required to achieve it?

At this point it becomes important to think in terms of the long term game. It is widely recognized that we are entering a period of dramatic workforce disruption. Fewer and fewer people are going to stay in the same company for life. Similarly, many are likely to require constant upskilling and reskilling as existing job roles are automated and new job types are created. Indeed, this is one reason behind the emergence of new models education that feature modularized and highly flexible micro credentialing.

So we need ways in which we can identify people’s skills and competencies, potentially gained in many different places, and then clearly map them to new roles, which could be in very different industries. We also need ways in which we can make it easy for people to identify paths to new careers and opportunities. That way they get just the training they need to transition into a new job. (You should not have to complete a whole new degree every time your job disappears!) Many people will have life skills and experience that could be invaluable in a new job role, but they might not be aware of how these enhance their existing skill set. For example, keen bushwalker, who has spent 20 years teaching new club members to find their way in wilderness areas has an essential leadership skill that might not be recognized in their current job role as an accountant. However, these extra informal skill sets might make them a prime candidate for moving into a role as a data scientist… or something else. We need ways in which we can make the invisible visible, to students, employees, educators and employers.

If you have ever tried to claim prior credit in a higher education institution, then you are probably aware of what a thorny problem the recognition of prior learning can be. Is an Operations Research 3 course completed at one institution equivalent to the Stochastic Operations Research course that is offered somewhere else? Does 3 years of experience as a project manager mean that someone entering a Masters degree can skip the Project Management course that it considers a pre-requisite for later courses? Is an Australian student who has achieved first class honors in their Bachelors of Science degree equivalent to one in the US who has completed a Masters by research? Translating qualifications between institutions and different geographical domains is a highly challenging problem.

Data interoperability can help here. In this coming age of workforce disruption, we need educational data to make sense both syntactically and semantically, in all of the different domains where it might be useful. Standards such as xAPI make syntactic interoperability very achievable, but semantic interoperability will be a far harder challenge. For starters we will need to undertake a rather intensive mapping exercise to understand what precisely a course, unit, subject degree etc. is in the many different places where these terms are used. Many countries are already well on the way to achieving this. Many countries now have a qualifications framework, including Australia (the AQF), Scotland (the SCQF) and initiatives like the European Quality Assurance in Vocational Education and Training (EQAVET) program, we see many different countries working towards semantic data interoperability in the educational space. With concepts like profiles starting to really take hold in xAPI we see ways in which these different concepts can be contextualized and compared using e.g. SKOS hierarchies, which the Companion Specification to xAPI give us a great model for using.

As this work progresses there is a very real chance emerging that we will be able to use it to help people navigate workforce disruption. But for this we need to free educational data so that the people who create it through their interactions with our systems can use it elsewhere. It is socially irresponsible for separate institutions and workplaces to hold onto data about the skills, experience and increasing capabilities of the individuals who they are responsible for. Not just that, it will lead to poorer outcomes. It would mean that universities will have to devote increasing resources to establishing equivalency between degrees and extracurricular experiences. Similarly, workplaces who hold onto workforce educational data in a misguided attempt to stabilize their workforce leave themselves hostage to poor indicators of a potential employees skills and capabilities (e.g. GPA, and granting institution). And even worse, they will continue to provide poor training for the staff that they do have. Training that focuses upon compliance rather than actually filling knowledge gaps and improving capability. (Hands up – who is sick to death of their compulsory yearly fire safety online training torture?)

We can do better, but to do this we need to engage with the bigger picture. We need to create Personal Learning Record Stores that an individual can access for life. These need to be able to accept data from a wide variety of data sources… and they will need to be able to send data to even more. We need to make it easy for people to curate their learning data, and to create ePortfolios of their work that link to the underlying evidence stored in the PLRS. We need to be able to extract educationally relevant constructs from low level clickstream data, and to create sophisticated profiles of learners that help them to see their weaknesses, strengths and ways in which they could move to a new career. We need to help people make sense of a multitude of choices that will arise in increasingly fragmented and complex training/work environments. With genuinely interoperable educational data we would be able to do this and more. But this is a long term set of goals. A lot of research will need to be completed and we need to start that now. So, while we are firmly keeping our eyes on implementing xAPI profiles and building genuine interoperability between IMS Caliper and xAPI, we also need to be working towards this longer term agenda.

Exciting opportunities open up as we start working towards data interoperability. A PLRS should form a core component of a fully functioning University API, which would enable institutions to swap tech solutions in and out of their infrastructure as they become obsolete, and to open up various parts of their data warehouses to students, staff members, and the public so that they can help them to create truly innovative new solutions… No more being held hostage to poor solutions that we have no way of changing… but, like I said. It’s a long game we are playing here.

Luckily, we have some funding to help work towards these goals. I have recently advertised a fully funded PhD stipend to come and work with the Connected Intelligence Centre at the University of Technology Sydney (which is right in the heart of Sydney). Indeed, more projects than this are advertised, and if you go to this page then you will be able to find all of them, but if you navigate to project 5 (Data interoperability and analytics for lifelong personalized learning) then you will see the kind of questions that we are currently asking about how we could really harness concepts such as data interoperability, and to use it in creating increasingly personalized opportunities for lifelong learning.
We are also going to have a job opening up in this area soon… so if you are interested in working on an xAPI research project focussed on the long term game of data ownership for lifelong learning, then why don’t you contact me?

Kirsty Kitto
Senior Lecturer in Data Science, UTS
Director, DISC

Kirsty Kitto (@kirstykitto) is a Senior Lecturer at the University of Technology Sydney (UTS). She models the ways in which humans interact with complex information environments, paying special attention to the interdependencies between language, attitudes, memory and learning. She works in the Connected Intelligence Centre (CIC) where she is seeking new ways of using data to help people navigate an increasingly connected world. She is the International Coordinator for the LASI-Local Network. On the research front, she is currently leading a project funded by the Australian government which is developing xAPI based solutions for instructors who want to teach “in the wild” beyond the LMS, and a grant funded by Graduate Careers Australia which is seeking to use xAPI to use learning analytics to help university students work towards developing evidence about their skills and capabilities in a chosen career. In past roles, Kirsty has worked on many projects in partial secondments to QUTs Learning and Teaching Unit, including the Learning Futures project, the creation of a new generation of teaching performance metrics, and the REAL employability project. Kirsty is a founding member on the Board of Directors for the Data Interoperability Standards Consortium (DISC).

Speak Your Mind