Assessment in Aotearoa Classrooms
A critical reflection on Gavin Brown's lectures

In junior Digital Technologies, the central assessment issue is that the artefact lies. A program that runs tells me very little about who understands it: code can be copied, assembled from tutorials, or increasingly generated by AI, and group projects like robotics blur individual contribution further. Brown's insistence that assessment evidence must be verifiable — something a colleague could look at and reach the same conclusion — names a discipline Digital Technologies teachers already improvise: checkpoint evidence gathered across the build (versioned code, design journals) and short walkthrough conversations — "why is this loop here?" — that make an individual student's thinking visible. The product alone is never the assessment; the trail is.

Senior Social Studies has nearly the opposite problem: internals are so literacy-saturated that they can measure writing fluency as much as social-science understanding, under-assessing students who can think historically but cannot yet control academic English. Marking them is where reliability wobbles — Brown was blunt that "humans are the problem" in scoring, susceptible to fatigue and mood. Departments manage this with exemplar-anchored rubrics, cross-marking, and moderation conversations about where achieved ends and merit begins. Both of my areas now face a new layer: the Year 9–10 progress indicators, whose quantifiers ("some, many") Brown showed are too vague to discriminate — with no funded mechanism to build common teacher understanding. The studio-art panel moderation he described, teachers physically ordering work and negotiating boundaries, looks like the model worth stealing.

My key learning was Brown's chain of reasoning: every assessment is a sample of what we care about, therefore imperfect, therefore the real design question is validity — we must "test what we value, not value the stuff that's easy to test." In Digital Technologies the easy-to-test is whether the thing runs; in Social Studies it is Microsoft Word's squiggly-line layer of surface accuracy. Neither is the thing I value. His companion mantra — assessment exists to tell the teacher "who needs what next" — reframes assessment as diagnosis rather than verdict: a total score that only confirms the rank order a teacher already knows is administration, not assessment.

The significance is immediate. From this year I will be teaching Year 9–10 under twice-yearly SMART testing and five-level progress judgments, with the government owning the data and league tables a foreseeable consequence. Brown's warning that nothing ensures my "consolidating" matches another school's makes moderation a personal professional obligation, not an administrative chore. Through Te Kete Ako, the resource platform I am building, I have begun pairing resources with verbatim-verified curriculum statements and exemplar-anchored success criteria, because the SOLO taxonomy he recommended (Biggs & Collis, 1982) gives me a cleaner ladder for cognitive demand than quantity-based rubrics. And I will hold onto his hardest point: teachers must care about students and about truth — telling a student their best work is not yet good enough is painful, and doing it honestly is part of the job.