IB Follow Through Grading Explained: A Tutor’s Guide
If you tutor IB Mathematics, you’ve seen it: two students make the same early slip, yet one earns most of the marks while the other crashes to zero. That isn’t “luck” — it’s usually IB Follow Through grading (FT) and how it interacts with method marks and accuracy marks. Once you understand the logic, you can teach students to write working that earns the marks they deserve, and you can mark faster with fewer grey-area arguments.
This guide gives you a tutor-ready mental model for FT (often connected to the “own figure” idea), the situations where it applies, and the exact working habits that make follow-through possible.
IB Follow Through grading: what “FT” actually means
In many IB Mathematics markschemes, a later part is designed to build on an earlier result. When that happens, the markscheme may indicate that certain marks in the later part are awarded follow-through from the student’s earlier value. In plain English: if the student’s later method is correct using their own earlier result, they can still earn the dependent marks.
Tutors often describe this as the “own figure” principle: the student’s own figure is treated as the input for later work, provided the markscheme allows it and the later steps are logically consistent. The key word is “provided”. FT is not a promise that any later work will be credited — it’s a controlled way to avoid double-penalising a single early error.
Follow Through is not “free marks”. It’s the markscheme saying: “We’re judging the method, not punishing the same slip twice.”
Method marks, accuracy marks, and where Follow Through fits
IB marking conventions are designed to separate mathematical thinking from numerical correctness. You’ll often see this split into method marks (for a valid approach) and accuracy marks (for a correct result). Follow Through is a bridge between parts: it says which later marks can still be earned when earlier inputs are wrong.
| What you’re assessing | What tutors should look for | Typical FT implication |
|---|---|---|
| Method marks (M) | Correct structure: setup, transformations, valid reasoning | If the method is correct with the student’s own value, FT often preserves M marks |
| Accuracy marks (A) | Correct simplification, arithmetic, final value, units/format | Accuracy may still be lost, even if FT allows the route |
| Follow Through (FT) | Consistent use of an earlier result across dependent steps | Only applies where indicated; doesn’t override “wrong method” |
The tutor insight is simple: students don’t need to be perfect to score well — but they do need to be coherent. Clear, consistent working makes it obvious when a later method deserves follow-through.
When Follow Through is (and isn’t) awarded
A useful way to teach FT is to treat it as a dependency rule: “Does part (b) depend on part (a)?” If yes, the markscheme may allow follow-through from (a) to (b). If not, (b) is marked independently and an earlier mistake shouldn’t help or hurt.
FT is commonly available when…
- The later part explicitly uses a value found earlier (a parameter, gradient, probability, etc.).
- The student carries their own figure consistently and the later method is valid.
- The earlier error is a single slip (e.g., arithmetic) rather than a fundamentally wrong approach.
FT is commonly not available when…
- The later part requires a fresh method that doesn’t depend on the earlier result.
- The earlier error changes the meaning of the problem (e.g., wrong model/assumption).
- The student’s later steps contradict their own earlier result (inconsistent substitution or sign changes).
- The markscheme specifies “no follow through” for a particular mark (this varies by question).
In tutoring terms: FT is a tool for fairness, not a loophole. The same early error can lead to different outcomes depending on whether the later marks are structurally dependent and whether the student’s working makes that dependence visible.
A tutor’s checklist for FT-friendly working
The fastest way to help students benefit from follow-through is not to teach “FT tricks” — it’s to teach markscheme-readable working. Here’s a checklist you can reuse in lessons and in mock feedback.
- Label carried values. Write “Let a = …” or “Using my result from (a): …”.
- Keep exact values where possible. Avoid rounding until the end unless the question asks.
- Show substitutions. Don’t jump from a formula to a final number without the input line.
- Keep structure visible. Factorisation, completing the square, and key identities should be explicit.
- Track units and context. A correct method with inconsistent units often becomes unmarkable.
- Don’t overwrite. If a value is corrected, keep the earlier line so the dependency is clear.
- State the final answer clearly. Even when FT applies, final accuracy/format still matters.
Notice what’s missing: “Get the right answer.” That’s not the tutoring bar in an IB paper. The bar is: “Make your reasoning auditable.” When it is, method marks and follow-through become much more predictable.
Common Follow Through pitfalls that cost marks
1) Rounding too early
Early rounding can turn a correct method into an accuracy loss later — or make the follow-through chain ambiguous. A practical rule for students: keep exact fractions/surds (or calculator memory) until the final line unless the question demands decimal accuracy.
2) “Invisible” substitutions
Many marking disputes are just missing lines. If a student writes a formula and then a number with no substitution line, it’s hard to tell whether they used the correct relationship with their own figure. Tutors should insist on one extra line: the substituted expression.
3) Changing the problem mid-way
Follow-through can’t rescue a solution that switches models. For example, a probability part that assumes independence when the setup doesn’t (or vice versa) is a method issue, not an arithmetic slip. Train students to name assumptions explicitly when the question hinges on them.
4) Inconsistent carried values
The easiest FT loss is inconsistency: using one value in line 3 and a different value in line 4, or changing a sign without explanation. Teach students to box intermediate results and reference them.
How to turn FT into better tutoring (not just better marking)
The most effective use of IB Follow Through grading in tutoring is diagnostic. When a student loses marks, ask: “Was this a method error, an accuracy slip, or a visibility problem?” Visibility is the one tutors can fix quickly.
A simple routine for reviewing a mock script:
- Locate the first divergence. Find the earliest line that becomes wrong.
- Classify it. Method vs accuracy vs communication/notation.
- Check dependency. Which later parts depend on that value?
- Rewrite one solution. Have the student rewrite the solution with “carry labels” and substitutions.
Students learn faster when they see that one extra line of working can be worth multiple marks in a dependent chain — not because you’re gaming the markscheme, but because you’re making the mathematics legible.
How AI grading should handle Follow Through (and how Gradenza approaches it)
Follow-through is hard for automation for one reason: it’s stateful. A grader has to remember a student’s value from part (a), decide whether part (b) depends on it, and then evaluate the later method under a “use their own figure” condition. That’s not a single-line rubric — it’s a linked sequence.
Gradenza’s IB auto-grading is designed around this dependency structure: submissions (photo or Google Drive) are graded with an LLM using human-verified markschemes, and follow-through logic is applied statefully across question parts when the markscheme indicates it. When the model is uncertain, teachers get structured reports with amber-flagged items and one-tap overrides — so you can keep speed and professional judgment.
Practical note for tutors and departments: AI marking is most reliable when student work is legible and step boundaries are clear. Diagram-answer questions, for example, can still be graded by AI but are best treated as “informational” checks — Gradenza can flag them without forcing a mandatory teacher queue. The result is a workflow that stays fast even during mocks (inference cost is typically in the ~$0.007–$0.009 per question range).
Conclusion: make IB Follow Through grading work for your students
IB Follow Through grading rewards coherent mathematics. As a tutor, your job is to make “coherent” teachable: label carried values, show substitutions, avoid early rounding, and keep the structure visible. Students won’t just score better — they’ll also understand why a method earns credit even when a number goes wrong.
If you want to apply FT consistently at scale (without spending hours on manual cross-checking), see how Gradenza models follow-through across parts while keeping teachers in control. See how Gradenza handles FT grading.
