Beyond Pass or Fail: Understanding Results in xAPI

In traditional learning systems, outcomes are often reduced to simple checkboxes: “completed,” “passed,” or “failed.” But what if we want to understand more than that? What if we want to capture how well someone performed, how long it took, what they actually did, and what it all means?

This is where the Results object in xAPI becomes essential.

What Is the Results Object?

In an xAPI statement, the Results object tells us about the outcome of an activity — the evidence of what happened when a learner engaged with content.

It answers questions like:

• Did the learner succeed?
• What score did they receive?
• How long did they spend on the task?
• What was their answer or action?

Think of it as the part of the story that goes beyond “did they show up?” to “what happened when they did?”

Anatomy of a Results Object

The Results object includes several fields — most are optional, but when used well, they provide powerful insights.

Together, these create a richer, evidence-based picture of learning.

Why Does This Matter?

In decentralised and modern learning environments, we’re often pulling data from many places — video platforms, mobile apps, simulations, quizzes, even real-world assessments. The Results object acts as a common language to interpret outcomes across those varied systems.

✅ For learning designers:

Results help shape what good performance looks like — you can analyse not just if people completed an activity, but how well they did and where they struggled.

✅ For trainers and educators:

It provides insight into learner behaviour, gaps, and misconceptions — especially when response data is used.

✅ For platforms like Remote Reviewer:

Results fuel decision-making, automation, and personalised feedback. For example, sending follow-up content based on confidence levels or triggering peer review based on performance.

✅ For compliance and stakeholders:

It offers more than a pass/fail report — Results provide defensible, granular records of performance and achievement.

Real-World Example

Here’s a simple xAPI statement with a Results object:

"result": {
  "score": {
    "scaled": 0.85,
    "raw": 17,
    "min": 0,
    "max": 20
  },
  "success": true,
  "completion": true,
  "response": "17 out of 20 correct",
  "duration": "PT10M"
}

From just this one result, we know:

• The learner completed the quiz.
• They passed with a solid score.
• It took them 10 minutes.
• We even know how many questions they got right — not just a percentage.

This kind of detail becomes powerful when aggregated across cohorts, modules, or time.

Tips for Using Results Well

• Always include completion and success — they are key signals for both people and machines.
• Use response to log open text or multiple-choice answers — useful for feedback and error analysis.
• Don’t forget min and max when using raw scores — it provides context to interpret results properly.
• Use duration to explore engagement and time-on-task.

Common Pitfalls

• ❌ Relying only on completion = true without meaningful score or success data.
• ❌ Not including response, missing the chance to see what the learner actually did.
• ❌ Confusing success with completion. A learner might complete an activity but not succeed — and vice versa.

The Bigger Picture

In decentralised learning ecosystems, xAPI Results help us move away from siloed tracking and LMS dependence. They enable learning experiences to happen anywhere — and still be measured meaningfully.

If you’re building a learning ecosystem that spans tools, contexts, or even time zones, Results are one of your strongest allies for turning raw activity into real insight.

Final Thought

xAPI lets us capture the what and how well, not just the whether. That’s what makes it such a powerful tool for modern, flexible learning.

Next time you’re designing or reviewing learning data, ask yourself:
“What story does the result tell?” – If it’s just “complete,” you’re missing the best parts.

Let’s Talk

Curious how xAPI could support your learning programs — whether you’re big, small, or somewhere in between? Let’s have a conversation.

Written in collaboration with AI