Put two rubrics on the same spoken answer. A speaking rubric marks pronunciation and fluency, your own marking scheme marks the content, and one recording comes back with both.
O-Level Situational Writing carries 30 marks: Task Fulfilment 10, Language 20. The band descriptors, the word count, and why Band 3 to 4 is the real gap.
An AI marker on its own agrees with a human about a third of the time. What makes it useful is where the teacher's judgement goes in, and if it stays.
Teachers say Gen Alpha are harder to teach. The attention evidence is thinner than it sounds, and the workload numbers point somewhere other than marking.
Edit or delete any annotation, re-grade only when you choose, undo any change in one click, and see which skills each piece of writing shows.
Almost every AI education pitch cites Bloom's two sigma. The study ran three weeks and has never been replicated. What personalised learning can really do.
In the largest meta-analysis of feedback, 38% of interventions made performance worse. What that means for AI grading, and for the rubric behind it.
Seven weeks to the written papers, one to the Oral. A PSLE 2026 revision timetable planned backwards from the real exam dates, and what to give up on.
Read the class average row before any individual child. A dip down a column is usually the task, not the class. Why completion and attainment go together.
Thirty scripts, one rubric, and a first pass that is already done when you sit down. What actually changes when the marking is drafted before you start.
Forty essays can swallow a weekend. What changes when the first pass is drafted before you sit down, and why the marking still has to be yours.
