Kaa System Guide

Provenance (written 2026-09-29). Primary sources: research/veridian/WEC14-verified-facts.md (read in full — its "Assessment structure" section, and the 4.3.2, 4.3.4 and 4.3.6 topic sections used for the worked examples below); content/veridian/curriculum/wec14/paper-anatomy.ts (read in full — the source of this document's central load-bearing claim, quoted in §1); and International A Level Economics Spec.pdf (Pearson Edexcel International Advanced Subsidiary/Advanced Level in Economics, Specification Issue 2, June 2018), independently re-extracted for this pass with pdftotext -layout directly from a local copy, specifically pp. 50–51 (the assessment-objectives table and the Unit 1–4 AO-weighting table) — checked to confirm the figures already in the facts bank, not taken on trust from them. The specification PDF was also searched directly for the string "KAA": zero matches, confirmed independently rather than assumed. Inside content/veridian/technique-hub/wec14/, the following were read before writing, specifically to scope this document away from what they already do: forge-level-threshold-guide.md (the KAA/Evaluation level-boundary pairs, Argentina Oct 2021 Q10 and Italy June 2025 Q10), forge-chains-guide.md (the generic KAA/Evaluation level-descriptor prose, quoted verbatim from January 2025's mark scheme), forge-evaluations-guide.md (AO4's own map across the paper, and the Q7(e)-vs-Section-C descriptor-text divergence), forge-tariff-4-mark-explain.md, forge-tariff-6-mark-analyse.md and forge-tariff-2-mark-calculation.md (the per-tariff K/AP/AN mark tables). None of those documents separates Knowledge (AO1) from Application (AO2) from Analysis (AO3) as three distinct things in their own right — all of them treat "KAA" as one combined unit set against "Evaluation" — and that specific layer is this document's whole job. Every mark-scheme fragment quoted below is under 15 words; longer indicative-content lists are paraphrased. Not affiliated with or endorsed by Pearson Edexcel.

#WEC14 KAA System Guide — Knowledge vs Application vs Analysis, Not Just "KAA vs Evaluation"

#0. What this document adds, and what it deliberately doesn't redo

Every other document in this hub that touches "KAA" treats it as one thing set against Evaluation — a combined band with its own level ladder (forge-level-threshold-guide.md, forge-chains-guide.md) or a combined band whose boundary with AO4 is the object of study (forge-evaluations-guide.md). That is a faithful reading of most of the paper: on Section B's Q7(e) and on every Section C essay, the mark scheme genuinely does not separate Knowledge from Application from Analysis at all — it bands the whole response, as a whole, against one KAA descriptor written entirely in the language of "chains of reasoning," and that descriptor never once uses the words "Knowledge," "Application" or "Analysis" (forge-chains-guide.md §1 quotes it verbatim; this document takes that as established and does not requote it).

But on the rest of the paper — Section B's four smaller parts, worth 2 + 4 + 6 + 8 = 20 of the 80 marks — Knowledge, Application and Analysis genuinely are three separate, individually ticked marks, not one blended judgement. This document's job is that layer: what actually distinguishes a Knowledge mark from an Application mark from an Analysis mark where the mark scheme ticks them separately, with short, real, attributed examples — and, honestly, where the evidence for a clean three-way split runs out.

#1. The real AO structure, confirmed directly — and why "KAA" is the right word to use

The specification (Issue 2, June 2018, p. 50–51) does publish an official four-part assessment-objective structure, independently re-checked for this pass against a local copy of the PDF:

• AO1: "Demonstrate knowledge of terms, concepts, theories and models" — showing understanding of how economic agents behave.
• AO2: "Apply knowledge and understanding to various economic contexts."
• AO3: "Analyse issues and evidence, showing an understanding of their impact on economic agents."
• AO4: "Evaluate economic arguments and use appropriate evidence to support informed judgements."

The spec's own AO-weighting table (p. 51) gives Unit 4's four AOs as 4.7% / 5.6% / 7.2% / 7.5% of the full IAL — rescaled within the unit alone (divide each by the unit's own 25% share) to AO1 18.8% · AO2 22.4% · AO3 28.8% · AO4 30.0%, both figures independently re-verified against the PDF for this pass and matching WEC14-verified-facts.md's own prior extraction exactly.

But that table is a whole-unit design target, not a label a candidate or a marker ever sees attached to an individual question. paper-anatomy.ts states this directly, having already caught and removed a fabricated AO-label reintroduction on a different WEC14 paper in this course's material: "every Section B and Section C mark-scheme line read for this page is labelled with Knowledge/Application/Analysis/Evaluation shorthand (K, AP, AN, EV) or a KAA/Evaluation band, never an AO-numbered tag." This document's own independent search of the spec PDF adds a second, negative confirmation: the string "KAA" does not appear anywhere in the specification at all — it is not spec vocabulary, and it is not this course's invented shorthand either. It is the mark schemes' own real, repeated term for the AO1+AO2+AO3 combined band, printed that way on the page — for example, WEC14-verified-facts.md's own transcription of the Section C mark split as "12 KAA... + 8 evaluation," and Section B's smaller parts labelled "Knowledge 2, Application 2" or "K2 AP2 AN2" with no AO number in sight. So: the assessment objectives really are AO1–4, exactly as the task brief anticipated they might not be — and the paper's own real, printed working vocabulary for the first three of them, combined, really is "KAA." Both things are true at once, at two different levels of the same paper.

#2. Where K, AP and AN are actually three separate marks, not one banded judgement

Section B's structure (forge-exam-technique-guide.md §2; paper-anatomy.ts) gives four small, flat-additive parts before Q7(e):

None of these four is levels-based. Each named element is its own tick-box: present or absent, worth its stated number of marks, with no partial credit for a well-written sentence that happens to combine two elements and no bonus for developing one element at length instead of hitting a second, separate one. This is the entire reason the 2-mark part can be "Application only" with zero Knowledge credit on offer at all (forge-tariff-2-mark-calculation.md) — there is no rule anywhere on this paper that a mark must contain some fixed mix of AOs; the mark scheme simply states, part by part, which AOs are on offer and how many marks each is worth.

Once Q7(e) or a Section C essay begins, this changes completely — and not just in degree. forge-chains-guide.md §1 quotes the generic KAA descriptor for both, word for word, from January 2025's mark scheme, and it is worth noting explicitly here what that quote does not contain: the words "Knowledge," "Application" and "Analysis" do not appear anywhere inside it. Level 1 is "isolated, superficial or imprecise knowledge and understanding" with "no chains of reasoning"; Level 4 is "logical and multi-stage chains of reasoning." A marker reading a Section C essay is not, at any point, asking "is this sentence Knowledge, Application or Analysis" — they are asking how many stages a chain runs to. The three-letter breakdown is real, and it is genuinely what the combined mark is testing in aggregate (it maps directly onto the same three AOs from §1), but it is not a checklist applied sentence by sentence past Section B's smaller parts. The rest of this document is therefore scoped deliberately to the one place the three-way split is a real, checkable, sentence-level mechanic: Section B's 4-, 6- and 8-mark parts.

#3. What actually separates a Knowledge clause from an Application clause from an Analysis clause

Built from real, attributed mark-scheme content, not a general economics-teaching convention:

Knowledge (AO1) — true independent of the specific source booklet. A definition, named theory or general causal mechanism that would be equally correct if the question used a different country or a different number entirely.

• "Free trade between member countries (1) with a common external tariff on imported goods outside the region/bloc (1)" — January 2022 mark scheme, Q7(c), the two-mark definition of a customs union. Both halves are pure Knowledge: nothing about Turkey (that series' source country) makes either clause true or false.
• "Primary product dependency (the Prebisch-Singer hypothesis): countries that export commodities would be able to import less for given volume of exports" — January 2022 mark scheme, Q10. A named theory, stated in general form, before any real country is attached to it.

Application (AO2) — true only because of this specific source booklet's own data or context. A real number, a real country, a real named policy — the general point from above, now tied to the one the question is actually about.

• January 2022's own Q10, same essay: "In 2020 Pakistan had a Human Development Index (HDI) score of 0.60 whereas Argentina had a HDI score of 0.85" — a real, quoted figure that could not be true of a different pair of countries; this is what turns the generic HDI concept into a mark on this specific question.
• The single cleanest illustration on the whole paper that Application does not require any Knowledge alongside it: Section B's 2-mark calculation part is tariffed "Application 2" with no Knowledge component at all, confirmed in two separate series (forge-tariff-2-mark-calculation.md) — correct arithmetic performed on the source booklet's own real figures earns full marks with no definitional sentence required anywhere near it.

Analysis (AO3) — the causal or explanatory link that connects a Knowledge point to a consequence. Not a new fact and not a new number; the word "because," "which," "by" or "so that" doing real work between two ideas already on the page.

• The one place this pattern is visible with Pearson's own inline tagging rather than a tutor's inference: the Summer 2025 mark scheme (Publications Code WEC14_01_2506_MS), Q7(d), 6 marks, "analyse two roles of the World Trade Organization" — a China/RCEP source booklet. Three role-statements, each explicitly split into a Knowledge clause and an Analysis clause: "To promote free trade/trade liberalisation (1K) by reducing trade barriers... (1AN)"; "To set the rules of trade (1K) because greater trade flows could increase economic growth... (1AN)"; "To resolve trade disputes (1K) by providing a forum for members to negotiate trade agreements (1AN)." Each Knowledge half names what the WTO does, in general; each Analysis half is the one-step "by/because" explanation of why that role matters, not a second fact.

Where the evidence runs out, stated rather than smoothed over. This same Q7(d) is tariffed, per the mark scheme's own summary line quoted in WEC14-verified-facts.md, as "Knowledge 2/Application 2/Analysis 2" — six marks across three AOs. The three quoted role-bullets above each carry an explicit inline (1K) and (1AN) tag; none carries an explicit (1AP) tag in any extraction this course has made of this question. The most plausible reading, consistent with how every other Section B part on this paper is built, is that the 2 Application marks attach to the question's own real context — naming or using the "28 free-trade agreements"/RCEP detail the source booklet supplies — rather than to a fourth generic role. But that is this document's own inference from the paper's general pattern, not a quoted Pearson tag, and it is stated as exactly that rather than as a confirmed finding.

#4. Income inequality — the same three-way split, one topic further from trade

Applying the same test to a different spec area, to show it is not a trade-topic idiosyncrasy. January 2024's mark scheme (Q8, income inequality within a developed country, Sweden/USA context) lists its KAA indicative content as a set of possible causes (WEC14-verified-facts.md, full bullet list). Read against the definitions in §3:

• Knowledge: "wage differentials between the unskilled and skilled workers have been increasing" — true as a general labour-market mechanism, independent of Sweden or the USA specifically.
• Application: the question's own real anchor data, "Between 2007 and 2022 income inequality increased in both Sweden and the USA" — the specific fact that makes this essay about those two countries rather than inequality in general.
• Analysis: the mark scheme's own linking clause on a neighbouring bullet, "globalisation... the demand for, and wages of, unskilled workers have fallen relative to those of highly skilled workers in developed economies" — a because-clause connecting a cause (globalisation) to a stated, directional consequence (falling relative unskilled wages), the same one-step link the WTO example makes in §3.

This is this document's own classification of real, quoted mark-scheme content against the AO1/AO2/AO3 definitions in §1 — Pearson's indicative-content bullets here are not individually tagged (K)/(AP)/(AN) the way the rarer WTO example is — so it should be read as an application of a verified method, not as a second instance of Pearson's own inline tagging.

#5. A three-question test, built from §§3–4

Not a Pearson rule — original synthesis for this document, offered as a practical check on Section B's four smaller parts specifically (§2 established why it does not transfer to Q7(e) or Section C):

1. Would this sentence still be true if the source booklet used a different country? If yes, it is doing Knowledge work.
2. Does this sentence depend on a specific number, name or policy this question's own source booklet supplies? If yes, it is doing Application work.
3. Does this sentence exist only to connect two of the above — a "because," "by" or "which" joining a cause to a consequence? If yes, it is doing Analysis work.

A sentence that fails all three — a restated question stem, or a general observation with no mechanism and no data — is doing none of the three, which is exactly the gap forge-error-pattern-tracker.md and signal-examiner-intelligence.md document from the Evaluation side of the paper and this document confirms exists on the Knowledge/Application/Analysis side too.

#6. Sources and honest gaps

Directly verified this pass: the AO1–4 definitions and the 4.7/5.6/7.2/7.5% (rescaled 18.8/22.4/28.8/30.0%) weighting figures, re-extracted from a local copy of the specification PDF (pp. 50–51) independently of WEC14-verified-facts.md's own prior extraction, and matching it exactly; a direct text search of that same PDF for "KAA," returning zero matches. The customs-union, Prebisch-Singer, HDI and income-inequality quotations are taken from WEC14-verified-facts.md's own verified quotations of the January 2022 and January 2024 mark schemes; the WTO Q7(d) inline-tagged quotation is taken from the same file's own Summer 2025 addendum, itself independently cross-checked there against pdftotext -layout and -raw on the primary PDF.

Genuine gaps, stated rather than forced: the Summer 2025 series is not part of this course's 14-series fully-read archive — only Q7(d) (and, separately, Q7(a)/(b) and Q10) of that mark scheme have been checked, so nothing here claims the inline (K)/(AN) tagging style is Pearson's general practice; it is confirmed for exactly one question, and §3 says so. No second instance of inline AO-clause tagging was found anywhere else in the archive searched for this pass — whether that is because the style is rare, or because this pass simply didn't find a second one, is not established, and is stated as an open question rather than resolved either way. No likelihood percentage, grade boundary, or "confirmed in every series" claim is made anywhere in this document.