xAPI + BI: Analyzing xAPI Data With Business Intelligence Tools

When I speak with professionals looking to start utilizing the xAPI specification I try to always illustrate it as a two step process.

1. First, connect your platforms so that you have conformant xAPI data standardized and centralized in a Learning Record Store.
2. Secondly, and only after the first step is complete, can you take full advantage of that data. Data for data's sake is not helpful to anyone, you need to make your data actionable and useful for your particular needs.
   

After step one, how do I accomplish step two to make my data actionable? Typically, this involves some analytics, data visualization, and data processing. Some LRSs are built with front-end visualization and data exploration tools but sometimes that is not enough for your specific metrics or questions. But since xAPI is portable you can move it into any other platform; any Business Intelligence (BI) platform, data visualization tool, query engine, statistical modeling software, anything! xAPI can be seamlessly integrated into your already defined BI workflows as well. Gone are the days that data lived within a large enterprise platform where you are forced into one view of the data unable to manipulate or pull the data out of their black box. To paraphrase a popular advertisement, "it's your data, use it where you need it!"

To demonstrate xAPI's portability, I have taken data from the Yet LRS and ported it over to Microsoft's Power BI tool in order to further investigate and visualize that data. Below are the steps I took to make that happen.

Obtain Your Data From the LRS

When using the Yet Analytics LRS, even in the Free Trial, you can download a CSV file of all of the statements in the LRS in an instant. You can add xAPI properties in the activity stream by clicking the + icon, see the screenshot below.

Then once you have the properties you want in your export, click the arrow button to download your data. Voila, you now have all your LRS data in one CSV file ready to be ported to any other platform.

Port Your Data Into Another Tool

Now, let's take that data to the Microsoft's Power BI free trial. The Microsoft Power BI platform is a very popular business intelligence tool. It offers customization and data visualization for specific and complex queries. In order to bring in the CSV file, go to the left menu and click 'My Workspace' and then hit click '+Create' > 'Dataset.' Follow the prompts and there you have it, xAPI data in your Power BI platform.

Dive Into Your Data

Start exploring! There are several tutorials available for the Power BI platform. Below are a few graphs made utilizing the available data tools in the Microsoft platform with our Kokea Concepts data.

This graph is a break down of statements based on the platform the learner was utilizing. With this kind of view, we can quickly see platform adoption, and see that most learning is happening within the Kokea LMS.

This graph represents all of our Kokea employees broken down by Verb. The yellow is 'failed' so we can quickly see who may be having some trouble through their learning. Some learners in need of intervention may be Britney Benzel and Terrance Martins.

This graph represents score averages based on region, which we can see that we have some people who are not assigned to a region in our data. Better double check that one! But also we see that the North East team is scoring very well while the South West team seems to be struggling.

We can also filter the data by which platform it is coming from, as Statement > Context > Context-Platform in xAPI speak. In the above graph, we are looking at statements specifically in the Kokea LMS broken down by Verb. In the LMS, the majority of interactions are 'viewing' and 'answering.'

Taking a look at statements from the Kokea Vue platform, we see a different break down of Verbs. Kokea Vue is a self-serve video knowledge base so we can see our learners are 'watching,' 'launching,' and 'searching' most frequently.

These visualizations are only scratching the surface of what you can do with your xAPI data in different business intelligence or data visualization tools. To answer more specific questions about your learning data and use a platform you may already know, you can even manipulate the data in Excel or Google Sheets. I would encourage anyone with xAPI data to investigate some of the robust data platforms available to see your data a different way. After all, "it's your data, use it where you need it!"