Different graphs are good at representing different kinds of data structures – frequency, change over time, patterns of performance.
One of our favorite graphs for experience data is the network graph. There’s a reason for that – data stored as xAPI statements is structured like a directed network graph.
Here’s a single statement:
In this visualization, the person is in yellow, their action is red, and the thing they’re interacting with is in blue. The network graph shows the relationship between these parts of an individual statement of experience in a directed flow.
Seems simple enough, right?
But network graphs can get overwhelming, especially when we start to look at larger datasets.
So let’s take a look at how the network graph is structured and what kind of patterns it’s good at representing, using a network graph representing our team’s Slack chat over the last 6 months.
Team Collaboration over Time
First let’s scope our view down and take a look at a series of snapshots.
These three graphs each show a week of data, for three weeks in succession.
In Week One, we’ve got two to three main groupings of communication, with each team member clearly communicating with at least one other team member.
In Week Two, we’ve got two fairly isolated team members – one with a lot of activity, and one with barely any.
In Week Three, our core team has more shared connections, showing a lot of collaboration across a larger number of team members. Two outliers appear to be totally unengaged with their peers, however.
The differences in collaboration over time could be due to any of a number of things – different projects, vacation, different habits of work. The network graph is particularly good at showing connections and patterns of collaboration, giving you insight to how your team members work together.
High Level Patterns of Engagement
Let’s try that big view again, now that we’re oriented a bit more to the larger shapes the network graph makes.
Remembering that yellow are people and blue are the things they do, we can start to see some high level patterns.
Our network graph uses size to convey frequency. Our most active team members on our Slack channels are the biggest yellow dots. The content most engaged with on Slack are the big blue dots. And the biggest arrows between team members represent the highest volume of communication between individuals.
So at a glance we’ve got who’s most active, what content is most popular, and who’s collaborating most often, all in one graph view.
Connecting Experience and Learning
One of the great benefits of collecting not just learning but experience data is that we can just as easily bring in informal learning and activity data alongside traditional learning and training.
The network graph is just one way to look at experience data – but a powerful one, capable of representing not just the answer to a specific question, but the actual shape of the data itself.
Want to explore an interactive network graph? Check out our demo LRS and interactive analytics.