Reducing feedback variability in summative assessment: grade drivers, variability hotspots, evidence anchors.

Summative assessment feedback can vary in structure and evidencing even when rubrics are fixed. This variability makes it harder for students to identify what mattered most, and harder for staff to defend decisions during moderation. This lightning talk introduces a practical three point feedback framework that improves clarity without changing rubric criteria or marking standards.

The framework uses three ideas. Grade drivers are the small number of criteria that actually determined the band for a specific submission. Variability hotspots are criteria where judgement often varies or attracts student questions, so markers need to be explicit about the evidence relied on. Routine criteria remain in rubric ticks. Using this structure, markers write a short three point narrative: the strongest evidenced feature, the primary band-limiting issue (and if needed one secondary), and two to three concrete actions mapped to criteria. Evidence anchors are agreed in staff development so that comments reference observable proof points rather than impressions.
The talk briefly shows how this works inside Turnitin: rubric ticks provide criterion coverage, optional one line evidence notes support hotspots and boundary cases, and the three points explain the grade drivers. Delegates will leave with a simple template and checklist suitable for adoption in Computing modules and transferable to other disciplines.