A Star Gap Analysis

Provenance (2026-09-29). This is a synthesis/index document, not new research — no primary PDF was opened for this pass. It is built entirely from documents already in this technique-hub, read in full first: forge-level-threshold-guide.md (the worked just-misses/just-clears pairs marking every real level boundary on the Section C essay's KAA and Evaluation ladders), forge-error-pattern- tracker.md (the ten confirmed EP error categories, each with its own examiner-evidenced source), and forge-exam-technique-guide.md §4 (the danger-phrases/trap-taxonomy digest, including the paper-wide mechanics new to that document). Two further documents in this same technique-hub were read in full because they bear directly on the specific question this document asks — what separates a top-band answer from a strong-but-not-top one — and each adds a genuinely new, primary-source-checked finding neither of the first two documents states: forge-evaluations-guide.md (the mark scheme's own printed Level-descriptor text, extracted and diffed across three series, which finds that Q7(e)'s Evaluation Level 3 band omits the exact clause Section C's requires, and that Q7(e)'s KAA ladder skips Section C's "two-stage chain" Level 2 rung entirely) and forge-judgements-guide.md (the verbatim descriptor source of "an informed judgement," and the country-gate wording nuance in its §1). signal-frequency- and-boundaries.md §2 was checked for real grade-boundary data — see §0 below for why none of its figures are reused here. Every claim below carries the same citation, quote and series/question attribution its source document already gives it; nothing here is re-verified against a fresh primary source, and nothing here invents a diagnostic category, a percentage, or a boundary the source documents don't already state. Quotes over roughly 15 words are paraphrased rather than reproduced, even where a source document quotes them at greater length. Not affiliated with or endorsed by Pearson Edexcel.

#WEC14 — SIGNAL: A* Gap Analysis (Unit 4: Developments in the Global Economy)

What this is. A self-diagnostic tool: a set of questions to run against your own last timed answer on this paper, each one built from a real, already-cited finding about what specifically kept a real graded response short of the top band, or let it reach it. forge-error-pattern-tracker.md gives you a log to record that a mistake happened; this document exists to answer the question that log doesn't — why does this specific habit cap the mark, on this specific tariff, and what real evidence says so.

What this isn't. New research, a new taxonomy, or a grade-boundary document. Every diagnostic item below maps to a finding, boundary or trap that already exists in one of the five source documents named above — see the citation after each one.

---

#0. What "top band" means here, and the grade-boundary data this deliberately doesn't repeat

WEC14 does have real, sourced grade-boundary data: signal-frequency-and-boundaries.md §2 states, across ten series checked directly against Pearson's own published boundary documents, that the A* raw-mark boundary out of 80 has ranged 56–61 and the E (minimum-pass) boundary has ranged 26–36. That range is not repeated or converted into anything below, for the same reason the sibling WBS12 document gives: it answers a different question. A whole-paper cut-off tells you what a finished, 80-mark paper needed to bank as a grade; it says nothing about what separates the bottom of a level from its top on any one question, which is the thing you are actually doing while writing. This document is about the second kind of gap.

What "top band" means here instead is the real ceiling structure forge-level-threshold-guide.md §1 establishes for this specific paper: Level 4 (10–12/12) on a Section C essay's KAA half, Level 3 (7–8/8) on its Evaluation half, Level 3 (7–8/8) on Q7(e)'s own KAA half, Level 3 (5–6/6) on Q7(e)'s own Evaluation half, full marks on the flat additive tariffs (Section B's 2/4/6/8-mark parts), and clearing the country/development gate wherever the question stem carries one. "A good pass" means accurate, on-topic, mid-band work with no factual errors; "the gap" is the specific, named thing that answer is still missing — not a vague call for "more detail" or "more evaluation." Nothing below converts any of this into a raw-mark total across a whole paper.

---

#1. This paper has three genuinely different ceilings, not one marking system reused three sizes

forge-exam-technique-guide.md §2 and forge-evaluations-guide.md §1 both already establish this in full; stated briefly here because every section that follows depends on knowing which ceiling applies before diagnosing the gap against it.

A gap-diagnosis run against the wrong row here is worse than no diagnosis at all — checking a Section B 8-mark answer for "an informed judgement" (§3 below) is checking it against a requirement that specific tariff's own mark scheme does not carry.

---

#2. Section C essay — where the KAA gap and the Evaluation gap actually sit

The KAA ladder (12 marks, Level 1–4). forge-level-threshold-guide.md §2 works this boundary-by- boundary on a real anchor — October 2021's Q10, evaluating the disadvantages of a current-account deficit to a developing country of choice — with a full just-misses/just-clears pair at each of the three boundaries. That worked pair is not reproduced here; running it yourself is the point. What this document adds is the diagnostic question each boundary is actually testing:

• L1→L2 (3→4): Does a real country enter the answer, and does at least one point run a genuine two-stage chain, rather than staying generic and country-free?
• L2→L3 (6→7): Does that chain run a third stage, ending in a named variable moving in a stated direction — not "the economy suffers," but a specific rate, balance or indicator named and directed? This exact ceiling — a two-stage chain capping Level 2 regardless of how accurate the economics is — is confirmed in near-identical wording across four series and three different essay topics (forge-exam-technique-guide.md §4, item 1); it is also forge-error-pattern-tracker.md's own EP10.
• L3→L4 (9→10): Is there a second, genuinely distinct mechanism developed to the same standard — not the same single mechanism restated from a different angle?

The Evaluation ladder (8 marks, Level 1–3, no Level 4). forge-level-threshold-guide.md §3 does the same for this ladder, anchored on June 2025's Q10 (Italy's public expenditure). Its own boundaries:

• L1→L2 (3→4): Does a generic "it depends" become one real, named condition that would actually change the verdict?
• L2→L3 (6→7): Is a second condition tested directly against the first, closing on an actual stated verdict — not two conditions merely listed in parallel?

What "an informed judgement" actually is, verified against the mark scheme's own printed text, not just the examiner report's gloss on it. forge-judgements-guide.md §1 re-extracted the Section C Evaluation Level 3 descriptor directly: it recognises different viewpoints and is critical of the evidence, "leading to an informed judgement," with reference to evidence and a logical chain behind it — confirmed identically worded across three questions in one series and independently re-checked in a second series. That five-word clause is the real source of the phrase every examiner-report-based document quotes secondhand. The anatomy of what earns it — decide, weigh, test one condition against a second, and close on a verdict with its own stated limit — is built out in full, against a real, previously-unused question (January 2021's Q9, Yemen and the case for aid), in forge-judgements- guide.md §§2–3; that worked judgement is not reproduced here, to avoid duplicating a full exemplar this technique-hub already carries — read it there.

A fabrication risk specific to this essay type, not to Q7(e) or Section B. forge-judgements- guide.md §2 names it directly: a Section C essay carries no source booklet, so the country and the supporting fact both come from the candidate's own knowledge. The gap-relevant risk this creates is inventing a plausible-sounding statistic about the named country to dress up a condition as anchored, when it isn't. The discipline is the same one this document itself follows — name the real limit in what you actually know, rather than dress up a guess as a fact — just applied to your own answer instead of to a research pass.

---

#3. Q7(e) — a genuinely different ceiling from Section C's, not a smaller copy of it

This is the single most load-bearing finding for a document about the good-pass/top-band gap, because it means the Section C diagnosis in §2 does not transfer to Q7(e) unchanged — and until forge-evaluations-guide.md checked the mark scheme's own printed descriptor text directly, no source in this technique-hub had confirmed exactly where the two diverge.

On the Evaluation side. Q7(e)'s Level 1 and Level 2 Evaluation descriptors are word-for-word identical to Section C's own, confirmed across three series (forge-evaluations-guide.md §3): Level 1 is generic evaluative comment with no logical chain; Level 2 is evaluation of alternative approaches with a partially-developed chain. Level 3 is where they diverge. Q7(e)'s own Level 3 (5–6/6) asks for a chain that "recognises different viewpoints and/or is critical of the evidence," plus context and a logical chain — and stops there. Section C's Level 3 (7–8/8) opens with the identical clause but continues into "leading to an informed judgement." That four-word clause is present in every Section C Evaluation Level 3 descriptor checked and absent from every Q7(e) Evaluation Level 3 descriptor checked, across the same three series, checked twice. The practical consequence: a Q7(e) evaluative point can reach this paper's top Evaluation band for this tariff by being genuinely conditional and critical of the evidence — it does not need the closing "on balance, X, because Y" verdict Section C's own top band specifically requires. A verdict on Q7(e) still costs little to add and still evidences "critical of the evidence," but it is not the precondition for the ceiling here that it is in §2.

On the KAA side. Section C's own Level 2 (4–6/12) is built specifically around "only two-stage chains of reasoning" — the exact phrase behind §2's own L2→L3 finding. Checked directly against Q7(e)'s own KAA descriptor table across the same three series: that phrase does not appear anywhere in it. Q7(e)'s three-level KAA ladder instead reads as Section C's Level 1, then something close to Section C's own Level 3 ("chains of reasoning… not developed fully or some stages omitted"), then something close to Section C's own Level 4 ("logical and multi-stage chains of reasoning," "fully integrated" — with a small wording drift between the earliest series checked and the two later ones, noted rather than smoothed over). Q7(e) is best read as skipping Section C's own middle "two-stage chain" rung entirely, not compressing all four Section C rungs into three (forge-evaluations-guide.md §4). This directly resolves a gap forge-level-threshold-guide.md §5 states and explicitly declines to guess at — whether the two-stage-chain ceiling transfers to Q7(e)'s own KAA boundary. Checked directly against the primary descriptor text: it does not, because the mechanism the inference would have extended is not printed in Q7(e)'s table at all.

So what actually caps a good Q7(e) answer below its own ceiling, if not a missing verdict? The real, confirmed pattern is breadth without depth — trying to cover as many of the source's measures as possible without developing any one of them, named directly in two examiner reports four years apart (forge-exam-technique-guide.md §2). forge-level-threshold-guide.md §5 builds one real near-miss/near- clear pair on this exact failure mode, from October 2021's Zambia growth-constraints question; that pair is not reproduced here — see it there. The honest limit forge-level-threshold-guide.md §5 already states stands: whether the identical two-stage-chain ceiling governs Q7(e)'s own Level 1/2 KAA boundary was, at the time that document was written, an unconfirmed inference — forge-evaluations-guide.md §4 has since checked it directly and found the mechanism doesn't exist in Q7(e)'s table, which is itself the answer to that stated gap, not a new open question.

---

#4. Section B's 8-mark part — the ceiling here is additive, not a level at all

A third, distinct shape again. forge-evaluations-guide.md §2 extracts the mark scheme's own instruction line for this tariff's 2 Evaluation marks, confirmed across four series: up to 2 marks for evaluative comments, distributed either as one point worth both marks, or two separate, shallower points worth one mark each. There is no third mark, no level, and — unlike Q7(e) or Section C — no requirement that the two marks come from one connected chain. The real worked instance forge-evaluations-guide.md §2 gives (January 2024's Q7(c), Greece's international competitiveness) shows three separate credited bullets splitting the two marks three different ways, all valid.

The diagnostic question for this tariff is therefore not a level question at all: is there a second, genuinely separate evaluative idea on the page — distinct from the first, not the same point restated — or does the answer stop after one, leaving a mark unclaimed regardless of how well-developed that one point is? forge-judgements-guide.md §4 gives a real worked bullet at this exact tariff (January 2021's Q7(d), causes of income inequality): one sentence that both names a real condition and commits to an actual comparative verdict inside two marks, contrasted with the examiner report's own diagnosis of the failure mode — a candidate who states the condition and stops has one idea, not two, and the second mark goes unclaimed.

---

#5. The country/development gate — a hard override, not a ladder rung

Everything in §§2–4 describes a ladder or an additive tariff climbed by adding real content. The gate works differently: printed directly above the descriptor table, it is a ceiling imposed on top of the ladder, regardless of how well the ladder has otherwise been climbed. forge-level-threshold-guide.md §4 states the real mechanic: a country-free answer on a gated question is capped at "maximum of Level 3 (9 marks)" — confirmed near-identically across at least 12 of the 13 mark schemes read (forge-error-pattern-tracker.md EP5). Some essays stack two independent gates on one question — a missing developing country, and separately, arguing growth where development was actually asked for.

Applied as a top-band diagnosis, specifically: a Level-4-quality KAA argument — two genuinely distinct, fully developed mechanisms — earns 10–12/12 only if the named country runs through it throughout. The identical argument with the country deleted is not a slightly weaker Level 4 answer; the gate overwrites whatever mark the ladder alone would have produced, down to 9, in one step (forge-level-threshold-guide.md §4). This is the one gap on this paper where "the content is otherwise Level 4" and "the actual mark is capped at Level 3" can both be true at once, which is exactly why it is worth checking first, before running §2's own boundary questions on the same answer.

Not universal — check the stem, not the pattern. October 2023's Q8, worded generically about "a country's terms of trade" rather than "a country of your choice," carries no such gate at all (forge-level-threshold-guide.md §4). forge-judgements-guide.md §1 adds one further precision: the gate tracks the category the stem names (developed / developing / generic), not the literal presence of the words "of your choice" — a stem that simply says "a developing country," with no "of your choice," still carries the identical N.B. cap.

---

#6. Which danger phrases are specifically top-band ceilings, and which stop an answer scoring at all

forge-error-pattern-tracker.md names ten confirmed EP categories; forge-exam-technique-guide.md §4 adds paper-wide mechanics on top of the curriculum lessons' own 40-item trap-taxonomy. Most of that combined list is knowledge-level confusion — a reversed definition, a mixed-up pair of models — that caps a discrete mark near zero regardless of how the rest of the answer is written, not a gap between a good pass and the top band specifically. The subset below is the second kind: cases where the underlying economics is right, but a specific, nameable habit still keeps the answer off its actual ceiling.

Specifically top-band ceilings (content can be accurate and the answer is still capped):

• EP5 — country-gate omission. Caps a Section C essay's KAA at Level 3 regardless of ladder position (§5 above).
• EP10 — evaluation stalls at a two-stage chain. The KAA-side version of the same mechanism caps Level 2 on a Section C essay (§2 above) — AO4's single largest-weighted failure mode on this paper, given evaluation's 30.0% share of Unit 4 (forge-error-pattern-tracker.md EP10).
• EP9 — off-target/tangential paragraphs inside an otherwise relevant answer. Distinct from EP3 (answering the wrong category entirely): content that starts on-topic and drifts, inside work that is otherwise developed enough to be climbing the ladder.
• Two-stage chains cap Level 2 KAA, confirmed across four series and three topics (forge-exam-technique-guide.md §4, item 1) — the general mechanic EP10 and §2's own L2→L3 boundary both describe.
• Bullet-point structure caps the format itself, on Q7(e) as well as Section C — confirmed directly in one series (forge-exam-technique-guide.md §4, item 2): a connected, multi-stage chain cannot exist on the page if the answer is written as terminated bullets, whatever the underlying economics.
• Solutions are not evaluation, on Section B as well as Section C — confirmed on an 8-mark Section B part in the earliest live series, not only on 20-mark essays (forge-exam-technique-guide.md §4, item 6): a candidate who proposes a policy fix where the question asked for causes or effects earns nothing for that paragraph, however well-reasoned the proposed fix is.
• Copying the extract earns zero analysis, confirmed on Section B's 8-mark part and on Q7(e) across five years (forge-exam-technique-guide.md §4, item 7) — a real risk specifically for a well-prepared candidate, since the fact being copied is usually correct; the mark scheme credits what you do with a fact, not the fact itself.
• Level 3 Evaluation on a Section C essay needs a stated, informed judgement — not more evaluation points (forge-exam-technique-guide.md §4, item 8; verified against the mark scheme's own text in §2 above). Four well-written conditions with no closing verdict is not more likely to reach Level 3 than two conditions that close on one.

Content-level confusions (right vs wrong; caps a discrete mark or an MCQ near zero at any competence level, not specifically a top-band phenomenon): EP1 (percentage vs percentage-point), EP2 (paired-definition reversal), EP3 (causes ↔ effects/policies substitution), EP4 (two-model/two-institution conflation), EP6 (market-orientated vs interventionist misclassification), EP7 (diagram/data-reading direction errors), EP8 (definitional half-answer). Each of these is real and evidenced (forge-error-pattern-tracker.md Part 1), and each is worth fixing — but fixing one moves an answer from wrong to right on that specific mark, not from a good pass toward the ceiling on an otherwise-accurate answer, which is this document's own narrower purpose.

---

#7. Your own error-log, read as a top-band diagnosis

forge-error-pattern-tracker.md Parts 2–3 give you a log and a tally to record which EP code a mistake falls into. This table adds what that log doesn't state on its own: what each top-band-relevant category actually costs against the ceiling, not just that it happened.

EP1/EP2/EP3/EP4/EP6/EP7/EP8 are deliberately not repeated in this table — §6 above states why: they cap a discrete mark or an MCQ near zero at any competence level, which is a real and different kind of gap from the one this document is built to diagnose.

---

#8. What this document deliberately leaves out

• No grade boundary, raw-mark total, or likelihood figure appears anywhere above. WEC14 does have real, sourced boundary data — see §0 — but it answers a different question from the one this document asks, and none of its figures are repeated here.
• No new diagnostic category. Every checklist item in §§2–6 maps to a finding, boundary or trap already named and cited in forge-level-threshold-guide.md, forge-error-pattern-tracker.md, forge-exam-technique-guide.md §4, forge-judgements-guide.md or forge-evaluations-guide.md.
• The full worked just-misses/just-clears pairs are not reproduced here. forge-level-threshold- guide.md already has one pair per Section C boundary (Argentina, Italy) and one near-miss/near-clear pair for Q7(e) (Zambia); forge-judgements-guide.md already has one fully worked judgement (Yemen). This document points to each rather than duplicating it, exactly as the task briefing for this document requires.
• No claim that any pattern above is "confirmed in every series without exception." Every source document states its own real denominator (e.g. "12 of 13 mark schemes," "confirmed across three series checked"), and this document repeats those denominators exactly rather than rounding any of them up.
• No document-type-doesn't-apply disclaimer. Unlike a diagram-mark-scheme-required check, this paper's Section C essays and Q7(e) are genuinely levels-marked, so a top-band gap-analysis genuinely has real, sourced material to synthesise from — this document is not a placeholder.

---

#Sources

forge-level-threshold-guide.md (the worked KAA and Evaluation boundary pairs for Section C, and the Q7(e) near-miss/near-clear pair — §§2–3, 5; the country-gate mechanic — §5); forge-error-pattern- tracker.md (the ten EP categories with their own confirmed sources — §§2, 6–7); forge-exam-technique- guide.md §§2, 4 (tariff-by-tariff evaluation requirements, and the paper-wide danger-phrases digest — §§1, 3–4, 6); forge-judgements-guide.md §§1–2, 4 (the verbatim Section-C-versus-Q7(e) descriptor divergence on "an informed judgement," the country-gate wording nuance, and the Section B 8-mark worked evaluation bullet — §§2, 4, 5); forge-evaluations-guide.md §§1–4 (the AO4 marks-by-component map, the Section B 8-mark part's additive "(2+0 or 1+1)" mechanic, and the primary-source-checked Q7(e)-versus-Section-C Level-descriptor comparison on both the Evaluation and KAA ladders — §§1, 3–4); signal-frequency-and-boundaries.md §2 (checked for grade-boundary data — see §0 for why none of it is repeated). Every claim above carries the same citation, quote and series/question attribution its source document already gives it; no primary PDF was independently re-opened for this pass. Not affiliated with or endorsed by Pearson Edexcel.