Attach the question context
Assignments carry exam, course, topic, marks, and dependency information where available.
Markscheme-based grading
Markscheme-based AI grading for math teachers
Gradenza grades math work against the markscheme instead of treating AI as a final-answer checker. It reads handwritten submissions, applies method and accuracy decisions, labels review points, and keeps teachers in control before feedback or mastery updates are released.
Grading report
M/A/Rmethod, accuracy, reasoning
FTdependency-aware decisions
Reviewteacher-adjustable reports
Audience
Who this is for
Use this workflow when the marking guide matters as much as the student answer.
IB Math teachers who need method, accuracy, reasoning, and Follow Through marks.
Tutors who want feedback reports grounded in the markscheme.
Schools standardising grading across assignments and mock exams.
Teachers wary of black-box AI scoring and looking for reviewable decisions.
Workflow
How markscheme-based grading works
The markscheme, question structure, and student working stay connected through the grading report.
Read the student submission
Students submit handwritten photos, PDFs, stylus exports, or Drive files for grading.
Apply markscheme decisions
The report separates method, accuracy, reasoning, and dependent marks instead of collapsing everything into a score.
Review before release
Teachers inspect the report, adjust decisions, then release feedback or write grades back when ready.
Use cases
Concrete markscheme-based grading examples
Markscheme-based grading is most useful when students can earn credit through process.
Method mark after a numerical slip
Award method credit for a correct setup even if the final calculation contains an arithmetic error.
Follow Through dependency
Evaluate a later part against the student own figure when the markscheme allows it.
Mock exam moderation
Use labelled decisions and amber flags so teachers can focus moderation on uncertain cases.
Benefits
Why markscheme-based AI grading matters
Math teachers need grading that explains why a mark was awarded or withheld.
More defensible feedback
Students see which part of the markscheme their working satisfied or missed.
Fairer partial credit
The grading report can recognise correct reasoning even when the final answer is not correct.
Better analytics
Because each result is tied to question and topic context, finalised marks can feed mastery maps.
Proof and trust
What makes it reviewable
The system is designed to support teacher judgment rather than hide it.
Visible mark decisions
Reports show the grading rationale at question or part level.
Teacher overrides
Teachers can adjust marks before feedback reaches students.
Exam-aware metadata
Question context can include topic, subtopic, marks, course, and dependency data.
Related resources
Related Gradenza pages and guides
Continue into the adjacent workflow instead of treating this page as a dead end.
FAQ
Common questions
What is markscheme-based AI grading?
It means the AI evaluates student work against the marking guide, including partial credit and method decisions, instead of only comparing final answers.
Can it handle IB Follow Through marks?
Yes. When the question dependency is known, Gradenza can evaluate later work against the student carried value.
Can teachers change the AI grade?
Yes. The report is designed for teacher review and adjustment before feedback is released.
Next step
Grade from the markscheme, not just the answer
Try Gradenza with one markscheme-based math assignment and review the grading decisions before release.