Prove it Podcast

Your assessment data can tell you far more than who passed.

In this episode Robbie Burns, Head of Business Consulting at Cirrus Assessment, talks to Gary Allison, Director of Learning Applications at Spiirall by LCvista, about using assessment data to measure the impact of learning, and what to do with what you find.

They cover:

• What learning impact actually means
• Why compliance metrics are the baseline, not the learning strategy
• Building evidence in stages, from pre- and post-assessments to retention check-ins and on-the-job observation
• Writing questions that produce data you can trust
• Simple first steps for teams who want to start measuring impact

Download the free one-page guide, Five questions to ask when the pass rate looks good, and hear when new episodes are out: https://bit.ly/4hIc45U

Ask most organisations about assessment right now and they’ll talk about AI. Security, cheating, proctoring, tighter regulations. AI has become the centre of the story, and that’s exactly the problem, because this was never really a story about technology.

This session challenges the assumption behind all that effort: that AI is the problem to solve. It isn’t. AI has simply exposed weaknesses that were already there, and the question worth asking is not “how do we stop candidates using AI?” but “what evidence would genuinely convince us that someone is competent?”

Most assessments are built around an ideal candidate. Someone calm, confident, digitally fluent, familiar with the process and sitting in perfect conditions. That candidate doesn't exist. And designing for them has real consequences for everyone who does.

In this episode, Robbie Burns (Head of Business Consulting, Cirrus Assessment) is joined by Cristina Gilbert (Strategic Partnerships, Cirrus Assessment) and Hugh McNeela (ReadSpeaker) to challenge one of the most deeply embedded assumptions in assessment: that consistency and fairness are the same thing. They are not. A consistent exam is the same for everyone. A fair exam works for everyone.

What we cover:

The ideal candidate. Most assessments are designed around someone who doesn't really exist. We look at what that costs you.

Reliability vs validity. Why consistent delivery does not guarantee valid measurement.

Accessibility as a validity issue. When a barrier has nothing to do with the capability you are measuring, it changes what your assessment is actually testing.

Legitimate vs accidental difficulty. Some difficulty belongs in your assessment. Some has just always been there. Knowing the difference is the starting point.

You will come away with the Three Layers of Assessment Fairness, a practical framework covering access, validity and candidate experience, plus four design questions your team can put to work straight away.

Want the framework on paper? Download the Assessment Fairness Toolkit at cirrusassessment.com.

Prove It is brought to you by Cirrus Assessment, the online assessment platform trusted by awarding bodies, universities and professional testing organisations.