I've been getting a lot of calls recently regarding xAPI. And so it seemed like a good time to organize and make more available several of the free resources my team at Yet Analytics has put together to help folks get started with the technology. Feel free to take these tools for a spin, and as always — reach out if you have any questions.
Why xAPI (in less than 50 words)
The Experience API (xAPI) increases the ability to measure outcomes in learning and training. And because xAPI tracks event-based data, it's perfect both for keeping track of formative learning and for keeping track of the health and usability of your learning technology ecosystem.
This "Getting Started" guide is meant for people who already have xAPI data sources. We call these "Learning Record Providers". If you are still looking for providers, check out the ADL's list of xAPI Adopters.
Once you've got a Learning Record Provider (or two, or three, or many), you'll need a Learning Record Store (LRS) in which to validate and store your statements. Note: "validate" and "store". An LRS doesn't just store data. It also validates that the data is properly formatted as xAPI. Why is this important? Because if your data is not properly formatted, you won't see any of the interoperability benefits of xAPI. Meaning, your data could turn into a big mess and you'll likely wind up locked into a specific vendor.
To validate your data, you can connect your Learning Record Provider to the Sandbox LRS from Yet Analytics. It's free. And it passes the ADL's xAPI LRS Test Suite. (Be wary of anything claiming to be an LRS that's not on that list).
The Sandbox LRS comes with an admin dashboard. This lets you see what's in your LRS. It also helps you get some immediate insights including what learning content is being accessed most often (or least often) and by whom. The data cards themselves can be customized... and if you upgrade to the Pro LRS, you can explore every possible attribute of what xAPI can capture.
Dashboards (The Good, The Bad, The Custom)
While we're on the topic, it's important to understand what an admin dashboard is... and what it isn't.
The Good: An admin dashboard is meant to be a tool to help you to monitor xAPI data. Some views into this data stream will be useful from a business perspective, such as quickly being able to identify top performers or learners who are most engaged with the technologies in your learning ecosystem.
But... Pro Tip: Don't confuse an LRS with a Data Visualization Dashboard. True, many LRSs come with a dashboard. But there are extremely useful LRSs that don't have any dashboard at all (such as our Developer LRS which is intended to be a massively scalable enterprise xAPI resource that can plug into existing BI dashboard and data visualization tools like Tableau, Kibana, and Power BI). The point of the LRS is to validate and store xAPI data. The dashboard — any dashboard — is an application that can visualize and help make sense of what's in your data. They are not one-and-the-same.
The Bad (and The Custom): Often, to really get at the important stuff — like measurements of activity against organizational key performance indicators (KPIs) and correlations between learning content and performance — it is useful to build custom data visualizations. If you've got IT or data science resources, this is relatively easily done... though there are nuances to working with xAPI data that can cause headaches for the less experienced. If you don't have the resources on-staff, it's definitely worth your time to talk with someone who has done this before. Either way, I happen to know a team of xAPI nerds that could help you out. In fact, the Yet team is currently building out an open source Data Analytics and Visualization Environment with the ADL. And yes. It's called DAVE. Because acronyms.
Being Adaptable (it's a good thing)
Organizations tend to have a lot of learning data already sitting around in spreadsheets. Unfortunately, you can't just export that data into an LRS. But you can use Yet's free Adapter to transform csv data into xAPI. It's not that hard to do. And the Adapter gives you hints along the way to make it easier.
In the interest of supporting the growth of the entire xAPI community and ecosystem, the Adapter lets you export transformed xAPI data to any LRS. We're all friends here.
Going Full Nerd
Sometimes you — or someone on your team — just want(s) to go full nerd. For you all, I present... xAPI Schema. This is a statement validator that can come in handy if you are writing your own statements by hand or want to see what kinds of errors you can throw. Open source (with the works) is available on Yet's GitHub page.
There is no shortage of great free xAPI resources available. For a great summer read, I'd suggest downloading the IEEE LTSC TAGxAPI Technical Report on xAPI that came out last year. The work of over 50 professionals in the field, it provides historical background and use cases as well as a full xAPI implementation guide.
There is also a standardization effort underway at the IEEE to turn xAPI into a formal standard. Check out what the working group is up to — all meetings are open to the public. And note that this work is especially important as xAPI is now a major component of the DoD's own instructions regarding distributed learning.
Lastly, if you haven't done it already, I'd strongly recommend taking part in the xAPI Cohort led by Torrance Learning. Over 200 people are already signed up for this year's group. As you'll find, across industry, government, and academia, there are a host of people just like you who are starting out on a journey to implement and make use of xAPI to improve the business insights you can get from your learning data.