Technique Masterclasses

Pure Mathematics 1

Examiner Intelligence

Provenance: compiled 2026-09-29. Source is research/veridian/WMA11-verified-facts.md in full (§2–§4, §6–§7, and the Lesson-audit log's own primary-source re-extractions) — every quote below was already matched against a real Pearson PDF by that document before this pass; no new primary-source fetch was done here. This document was written only after reading the other four WMA11 technique-hub files in this folder (forge-exam-technique-guide.md, codex-formula-and-method-bank.md, signal-frequency-and-boundaries.md, forge-battle-card.md) plus forge-forty-eight-hour-checklist.md and forge-error-pattern-tracker.md, specifically to avoid restating what any of them already cover. None of those six documents asks how the report itself talks — what phrases recur, what tone recurs, what that implies about how a script is actually read. That is this document's only subject. Direct quotations below are kept under roughly fifteen words each, with the series and question attached; anything longer than that is paraphrased. No likelihood percentage or grade boundary is stated anywhere below — where the source itself states a number (e.g. an examiner report's own "<5%"), it is quoted as the report's own words, not converted into a course-wide claim. Not affiliated with or endorsed by Pearson Edexcel.

WMA11 — Examiner Intelligence

Pearson Edexcel IAL Mathematics, Unit 1: Pure Mathematics P1 — how the report itself talks

0. What this is

Every other document in this folder tells you what the examiners found — a trap, a mark allocation, a frequency count. This one asks a different question: across the series this course has actually read, does the report's own language repeat? Not the mathematics — the phrasing, the emphasis, the things the reports habitually do that a mark scheme alone would never tell you.

The answer, checked below, is yes, in a small number of specific ways — and those ways say something concrete about how a real script gets read: examiners are not grading "did you get the right answer," they are grading a sequence of checkable moves, and their prose is built to say exactly which move was missing. Six patterns below are grounded in two or more independently-read series and labelled CONFIRMED. Three more are real, quotable, and specific, but were only found once in this pass — those are labelled SINGLE FINDING and left that way rather than dressed up as a trend.

Denominator, stated up front (matching signal-frequency-and-boundaries.md's own convention): the underlying facts bank has a fully-read examiner report for five series — January 2019, October 2020, October 2021, January 2023, January 2025 — plus report text for three more read at a single-question depth: June 2022 (Q5 only), June 2023 (Q1–Q2 only), and June 2024 (Q2(ii) only, recovered during a lesson audit rather than the main research pass). That is eight series with genuine, quoted examiner-report prose behind them. October 2022 and January 2024 are confirmed-genuine mark schemes with no examiner report mined — they contribute nothing to this document, which is about report language specifically. A pattern below citing three or four of the eight is a real majority of the available evidence, not a cherry-picked pair; a pattern citing exactly two is stated as exactly that.

1. Confirmed patterns

1.1 Every diagnosed error is priced, not just described

A WMA11 report almost never says "candidates struggled with X" and leaves it there. It names the wrong move and then states, in the same sentence, the exact mark consequence — because the mark scheme is the thing being explained, and a report that didn't do this would be useless to a marker checking their own judgement against it.

Four independent series, five separate questions, one consistent habit: describe the failure, then immediately state its price. The implication for how a script is read: a marker is not forming an overall impression of a solution's quality, they are walking a checklist and reporting, line by line, which line was and wasn't ticked — and the report is written in that same idiom because that is genuinely how the marking happened.

1.2 A shared, graded vocabulary of frequency stands in for numbers

Across every series with a fully-read report, the same small set of hedge phrases recurs, and they are not interchangeable — they form a rough scale from smallest to largest: "a small number", "a few", "some candidates", "a number of candidates" / "many candidates", "a significant number", "the majority". This is not one report's house style; the exact same graded vocabulary appears in January 2019 ("a significant number of candidates", Q8 ER), October 2020 ("a small number of candidates", Q3(b)(i) ER), October 2021 ("a few candidates", Q6(a) ER), and January 2023 ("a small minority", Q3 ER). The identical phrase "a small minority" recurs independently in June 2024's Q2(ii) report — a second, differently-dated series reaching for the exact same words for the exact same rank on the scale (a sub-group smaller than "some" but real enough to name).

The one place this pass found the scale abandoned for an actual statistic is January 2023's Q9(a) (period of tan x): the report states "less than half of all candidates gave a correct answer" for that specific sub-part, inside an opening line for the whole question that separately notes fewer than 5% scored full marks and over half scored zero across all of Q9's four marks (Jan 2023 Q9 ER — the whole-question figure, not the one-mark part (a) alone; conflating the two was a real, since-fixed bug in this course's own trig-graphs-period-and-domain.ts, per the facts bank's Lesson-audit log). This single instance is flagged, not generalised: one case of numeric language replacing the usual hedge-word scale is not evidence that examiners switch registers whenever a result is unusually stark — it is one data point, worth noticing, not a rule.

1.3 The printed "show your working" warning is a named, recurring flashpoint

Several WMA11 questions carry a printed rubric instruction — solutions relying on calculator technology are not acceptable — and the reports treat violations of it as their own category of failure, distinct from getting the maths wrong. This is confirmed across three series:

What makes this a pattern rather than three unrelated notes is where it sits in the report each time: general comments, the section written for the whole cohort and the whole series, not buried in one question's own paragraph. Examiners are treating "did you show the required working" as a standing, exam-wide concern worth a top-level mention, not a one-question quirk.

1.4 The most common story shape is "reached the right intermediate value, then stopped"

This is the single most-repeated narrative structure in the corpus, appearing across substitution, differentiation, integration and coordinate geometry alike:

Four independent series, four different topics, one shape: the mathematics that was attempted is usually correct as far as it goes; the mark is lost to stopping one step before the thing the question actually asked for. This is the strongest single implication in this whole document about how a script is read: examiners are not crediting "shows understanding of the technique," they are crediting "produced the specific deliverable named in the question" — an intermediate value, however correct, is worth nothing on its own if the question asked for what comes after it.

1.5 A correct calculation is not credited without an explicit closing statement

Distinct from 1.4 (which is about finishing the calculation), this pattern is about finishing the argument. Two independent series show the same expectation: getting the right numbers is necessary but not sufficient — the answer has to be stated as an answer to the question that was actually asked, in words if the question required a conclusion.

The mark scheme's own "General Principles for Pure Mathematics Marking" corroborates this as more than a two-question coincidence: its "Answers without working" clause states plainly that the rubric may withhold full credit from a bare answer, and its own house convention is that the accuracy mark, where a conclusion is required, is written to depend on the calculation and the statement together, not the calculation alone.

1.6 The reports occasionally step outside diagnosis into direct advice — and say so themselves

Most examiner-report prose is retrospective ("candidates did X"). A smaller amount is explicitly forward-looking, prescribing what a candidate should do differently — and the mark scheme's own general guidance confirms this is a standing habit of the report series as a whole, not an occasional aside from one particular report:

Two independently-dated per-question instances (Jan 2019, Oct 2021), reinforced by the general guidance's own plural attribution to "recent examiners' reports" as a class. The implication: a report is not purely a post-mortem — it is also, deliberately, a teaching document, and the exam board's own general guidance treats that dual role as routine rather than exceptional.

2. Single findings — real, but not (yet) a pattern

These are genuine, verbatim, citable — and were found in exactly one series each in this pass. Per this document's own standing rule, a finding seen once is reported as a finding, not inflated into a trend.

3. What this implies for how you write an answer

Briefly, because the technique itself already has a home in forge-exam-technique-guide.md — this section only states what follows directly from §1 above, not a new technique layer:

4. Series ledger for this document

SeriesDepth of report actually quoted hereUsed in
January 2019Full report§1.2, §1.6, §2
October 2020Full report§1.2
October 2021Full report§1.1, §1.2, §1.4, §1.5, §1.6
January 2023Full report§1.1, §1.2, §1.3, §1.4, §1.5, §2
January 2025Full report§1.3, §1.4
June 2022Q5 only§1.1
June 2023Q1–Q2 only(checked; no report-language finding distinct from the above emerged from this narrow a read)
June 2024Q2(ii) only§1.2, §1.3
October 2022 / January 2024Mark scheme only, no report minedNot used — no report language exists in this pass's sources for either

5. Flagged gaps