grayscale photo of building near body of water

Governing How Survey Questions Relate to Concepts

Today we defined the work-to-be-done to address the need for a top-down schema of the survey. In my previous post, I described that the first swipe of work is modeling the cmi.interactions used in the survey. That work is going apace, so today let’s focus on how I’m thinking about slicing and dicing our first xAPI Profile.

Focus on the Value

I describe all work to be done in terms of the value needed.

“As a nurse,
in a hospital with multiple products,
I will want any explicit disclosures
about the knowledge and skills, for which I’d like support,
to be able to inform my experience with semantic links, which associate my survey responses with other content that may be relevant to my explicitly-stated needs.” 

Who, What, When, Where, Why, How – all in one sentence
– that’s how I write a Value Proposition

This one sentence sets the data collection boundaries for what gets tracked, and in what conditions, and informs a limited scope. The Casey-Fink survey has several areas of focus, but in this context, we’re just going to focus on clinical skills, for which we have some content that’s been identified by an actual subject matter expert as being a strong or direct match to support the particular skill. We’re only going to look at data collection on likerts where the clinical skill involved was also responded with lower user-confidence.

If you’re familiar with xAPI Profiles, what hints have I just given you? If you’re not, no worries, we’ll reference this again.

Anyway, groovy. That’s a whole bunch of work we don’t have to wait on to move forward.

That means the work ahead has a few steps. We’ve not used xAPI Profiles ourselves, and it will be some time before we might realize the value of making the semantic associations, but I’m betting on it gets us closer to certain points of integration or reference sooner. It provides a consistent framework and set of tools for us to practice our assertions, and that’s more than enough reason for us to be using it. But it also means it’s going to take a lot more time to do this than it looks like, because it’s the first time and there needs to be lots of time for questions and sometimes drawing and editing diagrams to answer questions.

What’s the Work?

The goal for xAPI Profile, at this point, is relating survey items to their order/hierarchy in the survey and other semantic concepts (like learning objectives, competencies and/or content). Considering where we’re at, to do that we need to…

  1. Determine a scheme for unique identifiers that relate to content and competencies.
  2. Model xAPI statements, statement templates and accompanying documentation that are needed to support the relationships we’re looking to support into a document.
  3. Test the model xAPI statements for conformance with cmi5, using the CATAPULT testing tool.
  4. Author relevant semantic concepts from the exemplar xAPI statements as an xAPI Profile: https://profiles.adlnet.gov/

Related


Comments

One response to “Governing How Survey Questions Relate to Concepts”

  1. […] planning and architecture things is way more up-front testing. You might’ve noticed in my last post and previously, I’ve made reference to wanting to test stuff against CATAPULT, as our […]