Context: why mock-exam review often fails
A marked GCSE maths mock returns a dense page of lost marks, but most students review it passively: they read the teacher’s comments, nod at the arithmetic slip, and file the paper away. The result is predictable. The same error type reappears in the next assessment because the correction never becomes a retrieval task. Duffspin treats the marked script as a diagnostic source rather than a verdict, and its value lies in the sequence that follows the red pen.
In a typical duffspin cycle, the student does not rewrite the whole question. They isolate one faulty step, name the misconception in plain language, and then answer a parallel item from memory after a short delay. The claim is narrow: error review works when it produces a new correct response under test-like conditions, not when it produces neat notes.
Development: the three moves of a duffspin review
1. Classify the error before correcting it
The first move is diagnostic. A wrong answer in algebra may be a sign error, a fraction operation, or a misread instruction; each cause needs a different repair. The student writes the cause in a single sentence, for example: ‘I expanded the bracket but dropped the negative sign on the second term.’ This step turns a vague sense of failure into a specific target. The takeaway is that the error label must be concrete enough to test later.
2. Drill the smallest repair
The second move is a short, focused drill. If the cause is dropping negative signs during expansion, the student completes four or five expansions with negatives, checks them, and stops. The drill is deliberately small because the goal is not to redo the entire topic but to repair the step that failed. In tutoring, this is where a duffspin routine earns its keep: the tutor can supply parallel examples immediately, watch the student work, and correct the process before the error fossilises.
3. Delay the re-test
The third move is a delayed re-test, usually after 24 to 48 hours. The student answers one or two fresh questions without notes, then compares the new response with the original error label. If the same cause appears, the repair was too shallow; if the response is correct but slow, the student needs fluency practice rather than more explanation. This delay separates recognition from recall.
Implementation: a 20-minute duffspin session
A practical session stays tight. The student selects one error from the marked paper, writes the cause, completes a five-item drill, and sets a re-test for the next day. The tutor’s role is to keep the drill parallel but not identical, so the student cannot memorise the original answer. The session ends with a single sentence in the error log, not a rewritten page.
- Minutes 0–3: choose one error and name its cause.
- Minutes 3–15: drill four to six parallel items with immediate feedback.
- Minutes 15–20: schedule the delayed re-test and record the error label.
Result: a shorter error log with better transfer
After four weeks of duffspin reviews, the useful measure is not the number of pages rewritten. It is the number of repeated error types. A student who began with eight recurring mistakes may finish with three, and those three are now easier to target in revision. The marked mock becomes a working document: each red mark either disappears from the next script or returns with a clearer cause. That is the practical result of the duffspin sequence—less cosmetic correction, more retrievable repair.