Technique Masterclasses

Pure Mathematics 4

Examiner Mindset Guide

Provenance: written 2026-09-29, grounded in research/veridian/WMA14-verified-facts.md §4 (mark- scheme conventions, verbatim, confirmed stable Jan 2021 → Jun 2024) and §5 (the per-topic examiner- report citations). Checked in full before writing, specifically to avoid duplicating what already exists for this paper: forge-exam-technique-guide.md §2 ("The M/A/B Mark-Coding System — Playing It Correctly") already covers the same mark-scheme mechanics from the candidate's own "how to bank marks" angle; signal-examiner-intelligence.md already covers the recurring rhetorical patterns in how reports are written (severity vocabulary, the report-entry shape, mark-cost pairing). Neither document asks this one's question: what is the actual marking process Pearson's examiners follow against this specific paper's scripts, and what does the real record show about what they check for first. That is this document's angle, and it is additive to both, not a restatement. Also checked, to confirm no overlap: forge-battle-card.md, forge-error-pattern-tracker.md, forge-level-threshold-guide.md, and the twelve lesson files under content/veridian/curriculum/wma14/. Every quotation below is under fifteen words or is a short, clearly-marked mark-scheme fragment, each individually attributed to a real series and question/ section; none is invented, and no likelihood, percentage or grade boundary appears anywhere in this document. Where the record runs out on a sub-question, this document says so and stops rather than padding — see §4.

WMA14 · Examiner Mindset Guide

What this document is, and isn't

This is not a restatement of the M/A/B definitions (forge-exam-technique-guide.md §2 already gives those, from the candidate's side, as "how to play the system correctly"). It is not a study of how examiner reports are phrased (signal-examiner-intelligence.md already does that, across five series). This document asks a narrower, more literal question: when a real WMA14 script lands in front of a real examiner, what are they actually doing — what marking model governs the paper, what do they check first, and what does the primary-source record show, in the examiners' own words, about how that checking plays out on real scripts. Every claim below traces to WMA14-verified-facts.md §4 or §5, cited (Month Year, Qn) or by mark-scheme section name. Nothing here was assumed; the marking model itself (§1 below) was checked against the primary source rather than taken as given, per this course's standing discipline after a prior batch produced fabricated material of exactly this shape.

1. The real marking model for this paper: discrete point-tallying, not holistic best-fit

This is checked, not assumed, because the two models genuinely coexist inside Pearson's own qualification family and it would be a real error to assume one from the other. The Business unit already documented in this course (WBS12) explicitly uses levels-based, holistic marking for its extended-response questions — a 20-mark Evaluate question is marked against four generic level descriptors (1–4, 5–8, 9–14, 15–20), with an examiner reading the whole answer and placing it in a level as a "best fit," not counting discrete points (research/veridian/WBS12-verified-facts.md). WMA14 does not do this anywhere. Its own "General Marking Guidance," present verbatim at the start of every mark scheme reviewed (Jan 2021 → Jun 2024, confirmed stable across that span), defines three mark types and nothing else:

That last line is the direct evidence against holistic marking for this paper: a levels-based scheme is, by construction, a subdivided band an examiner places a whole answer inside; WMA14's own guidance explicitly forbids exactly that kind of subdivision at the mark level, because every mark is meant to be a discrete, independently-awardable point against a named scheme line. Every worked mark-scheme anchor this course has independently verified for WMA14 — proof by contradiction's M1→A1→dM1→A1* ladder (Jun 2022 Q9), the foot-of-perpendicular M1→M1→A1→dM1 ladder (Oct 2021 Q7), the binomial B1→M1→A1→A1 ladder (Oct 2024 Q1) — is a literal point-by-point tally against named criteria, not a band judgement. There is no question type on this paper marked by best-fit, because there is no extended-writing or "supported judgement" question type on this paper at all: every item is a calculation, a derivation, or a proof with a specific, checkable target.

2. What examiners are trained to look for first

2.1 The method, before the accuracy

The scheme's own hierarchy makes this literal, not just a marking-order convention: an A mark "can only be awarded if the relevant method (M) marks have been earned." An examiner cannot assess whether an answer is correct until they have first decided whether a method was genuinely attempted — accuracy marks are structurally downstream of that decision, not a separate, parallel check. For calculus specifically, the scheme states the actual gate in one line each: "Differentiation — Power of at least one term decreased by 1 (xⁿ → xⁿ⁻¹). Integration — Power of at least one term increased by 1 (xⁿ → xⁿ⁺¹)." That is the literal first thing an examiner checks on a calculus line — not whether the final answer is right, but whether the exponent moved the correct direction at all.

2.2 A quoted formula is checked before the substitution is checked

The scheme's "General Principles for Pure Mathematics Marking" states this as a real marking instruction, not folklore: "the advice given in recent examiners' reports is that the formula should be quoted first... Method mark for quoting a correct formula and attempting to use it, even if there are small mistakes in the substitution of values." The practical consequence for how an examiner reads a script: a quoted formula lets the method mark be awarded and checked independently of the arithmetic that follows it. Where no formula is quoted, the same mark can still be found — "the method mark can be gained by implication from correct working with values" — but the scheme itself flags that route as more fragile: it "may be lost if there is any mistake in the working." An examiner is, in effect, looking for the formula on the page before they look at what was done with it.

2.3 On a "show that" line, the path is checked, not just the destination

Where a question prints its own target answer, the scheme's general policy on rounding and exactness sits alongside a separate, real finding from the record itself (see §3.1 below): examiners are reading for whether every line is a genuine forward derivation from the one before it, because a printed target makes it possible for a candidate to work backwards and disguise a genuine algebra error. This is not this document's inference — it is what the record shows an examiner actually catching, quoted in §3.1.

2.4 A method the scheme didn't anticipate gets escalated, not improvised

The January 2026 mark scheme for Q8 (a proof-by-contradiction item, 4p²−q²=46) explicitly credits three separate, structurally different valid routes to the same proof, then adds a real instruction to the marker beyond those three: "If you see any other proofs which appear to be credit worthy then send to review." This is a genuine, checkable piece of the marking process itself, distinct from anything an examiner's report later says about how candidates did: a marker encountering a fourth, unanticipated-but-valid method is not trusted to award marks on their own judgement alone — the scheme's own instruction is to escalate it for review, not to decide unilaterally. (This line is from the mark scheme's marking instructions, not from an examiner's report — flagged explicitly since §3 below is examiner-report material specifically, and the two are different document types with different authority.)

3. Three real, cited pieces of examiner-report commentary on how marking plays out in practice

3.1 Examiners are trained to catch a "show that" answer that was reverse-engineered

January 2021, Q10 (separable differential equation, particular solution from a boundary condition, target answer printed on the paper): "The given answer here persuaded many candidates to 'adjust' their working following obvious mistakes." This is the clearest single piece of evidence in the whole archive that a printed target answer is not simply matched against a final line — the report is describing markers who, reading the intermediate working, could identify where a genuine error occurred and see that a later line had been altered to reach the given target anyway. The mark-scheme consequence follows directly from §1's rules: since every mark is a discrete point against a named scheme line, an examiner reading a script line-by-line is well placed to catch exactly this, in a way a single "does the final line match" check could not.

3.2 A script using an unscheduled but valid method still gets full credit

October 2021, Q7(a) (finding the point on a line nearest a given point, then the distance): alongside describing the dominant wrong method (dotting the wrong vector, scoring 0), the report notes "Other methods were seen in part (a) including minimising the length of AX using either differentiation or by completing the square." Neither of those two alternative methods is the scheme's primary route (solving AX·d = 0), yet the report records them as legitimately credited routes, not deviations. This is direct evidence that an examiner's first task on an unfamiliar-looking script is not "does this match the model answer" but "is this a mathematically valid method that reaches the same result" — a genuinely different question, and the one the mark scheme's own M-mark definition ("knowing a method and attempting to apply it," §1 above) is built to answer.

3.3 A correct calculation with no stated justification is marked as unfinished, not as done

January 2024, Q8 (proof by contradiction that a cubic has no stationary points): "The majority successfully set up their initial assumption... However, many then simply commented that the equation formed had no solutions, giving no explanation as to why... and so gained no further credit." The examiner is not marking whether the algebra reached a true statement — it did — but whether the candidate stated, in words, why that statement constitutes the required contradiction. This is the practical, checkable version of §2.1's "method before accuracy" hierarchy applied to proof specifically: a correct derivation with an unstated conclusion is, on this evidence, marked as an incomplete method, not as a correct one missing a final flourish.

4. Flagged gaps — what this document does not claim

Sources used

research/veridian/WMA14-verified-facts.md §4 (General Marking Guidance and General Principles for Pure Mathematics Marking, verbatim, Jun 2024 mark scheme cross-checked against Jan 2021) and §5.1, §5.5, §5.6 (examiner-report quotations for Jan 2021 Q10, Oct 2021 Q7(a), Jan 2024 Q8, and the January 2026 Q8 mark-scheme "send to review" line). research/veridian/WBS12-verified-facts.md (levels-based marking confirmation, used only for the cross-subject contrast in §1). Checked for duplication and confirmed not overlapping: forge-exam-technique-guide.md, signal-examiner-intelligence.md, forge-battle-card.md, forge-error-pattern-tracker.md, forge-level-threshold-guide.md, and the twelve lesson files under content/veridian/curriculum/wma14/. No new primary Pearson PDF was fetched for this document; every quotation reuses a fragment the facts bank had already independently verified against the source PDF text.

Not affiliated with or endorsed by Pearson Edexcel.