Topic Brief: Show That Answer Discipline

#SIGNAL Topic Master Brief — "Show That" Answer Discipline (WST01, cross-cutting; spec 2.3/2.4/4.1/5.3/6.1)

Provenance (2026-09-29): built from research/veridian/WST01-verified-facts.md (grepped for "show that", "showing your working", "cso" and "A1*"), the live lesson content/veridian/curriculum/wst01/show-that-answer-discipline.ts, and the per-lesson audit-log entries inside four sibling curriculum files — expectation-and-variance-of-linear-transformations.ts, paper-anatomy.ts, discrete-random-variables-probability-function.ts, and outliers-boxplots-comparing-distributions.ts — where a real "show that" instance surfaced during that lesson's own audit. Cross-checked against codex-deep-dive-show-that-answer-discipline.md and signal-frequency-and-boundaries.md so this document does not restate either: two of the four question-level citations below (Jun 2024 Q2(b); Jan 2023 Q5(d)) do not appear in CODEX's own synthesis table, and this document's headline frequency count is deliberately stricter than CODEX's (§1). No new PDF fetches were made.

#1. Frequency — seen in 3 of 6 series checked, not 4

The facts bank has a fully-read Principal Examiner's Report for 6 of the 14 WST01 series with a verified mark scheme: Jun 2019, Jan 2021, Oct 2021, Jun 2022, Jan 2023, Jun 2024. A confirmed "show that" (or equivalent printed-target) instance is directly evidenced in 3 of those 6: Oct 2021, Jan 2023, Jun 2024.

This is narrower than the CODEX deep-dive's own "confirmed in 4 of the 6," which also counts Jan 2021 — but Jan 2021's contribution there is the converse case (an ordinary, non-"show that" part where a bare answer legitimately scores full marks, conditional on precision), not an instance of the discipline being tested or failed. That case is real and relevant background (§5), but it is not itself a "show that" citation, so this brief counts it separately rather than folding it into the headline frequency. Jun 2019 and Jun 2022 have no confirmed instance in either document.

#2. What was actually tested, by series

• Oct 2021 — three separate instances in one series. Q6(a) (3 marks; Normal distribution; show that P(H > 30.8) = 0.15): "many students wrote down the standardisation followed by 0.15 missing out the intermediate step and the accurate answer. Some students simply wrote down 0.1500… with no working at all." Q3(b) (interquartile range; show that the IQR equals a given value): "the main errors, since the answer was given, were making up two ages... which differed by 16 as their 'working'." Q3(c) ("show that there are 3 outliers"): "having gained the correct limits some did not list the 3 outliers in this part in order to show there are 3 outliers." Plus a general comment: "many students do not really understand that evidence has to be given for 'show that' questions."
• Jan 2023 — a mark-scheme structural citation, not a candidate-failure quote. Q5(d) is coded M1, A1ft, A1* (3 marks); its A1* note reads "answer is given so there must be a fully correct solution given with no errors" — verified against the real mark scheme (WST01_01_MS_2301) during paper-anatomy.ts's own audit; this citation is absent from the facts bank itself. Plus a general comment: "Students need to understand that a full method should be given for 'show that' questions."
• Jun 2024 — two separate question-level instances. Q4(c) (regression line; show that g = −42.3 + 0.722d, 3 s.f.): "too often students lost the final A mark as they failed to give values to the required degree of accuracy... Students should be encouraged... to give answers to at least one more decimal place than the given value." Q2(b) (discrete random variable; show that Var(X) = 4.14): a real "show that" instruction carrying an inferred 3 of Q2's 12 marks (the mark scheme states 7 of 12 test spec 5.3 across parts (b)–(c); part (c) is confirmed at 4) — but no candidate-failure commentary for this specific part survives in either source. Plus a general comment: "questions that ask a student to show something is true require all the steps in the working to be shown."

Two further citations are named but too thin to build into this frequency count: Jan 2023 Q3(a)/Q4(i)(a) ("algebra") and Jun 2024 Q3(c) ("coding") are each named once, in the facts bank's own cross-cutting synthesis paragraph, with no surviving verbatim quote anywhere else in either source.

#3. Mark-tariff pattern

Wherever a mark count is confirmed, this discipline attaches to a small, self-contained sub-part inside a larger multi-part question, never a whole standalone question: Q6(a), 3 marks (2 M + 1 A1cso); Q2(b), 3 marks inferred as above; Q5(d), 3 marks (M1, A1ft, A1*). Q3(b)/(c) and Q4(c) have no confirmed individual mark count in either source, so none is stated here. On the parts with a full breakdown, risk concentrates in the last mark specifically — an A1cso or A1* — not spread evenly: a correct method with an under-precise or unsupported final line loses one mark, not the whole part.

#4. Recurring context/angle across the confirmed series

This discipline is genuinely cross-cutting by spec section: confirmed in Section 2 (interpolation/ outliers), Section 4 (regression), Section 5 (discrete random variables) and Section 6 (Normal distribution) — four of the paper's six spec sections. It has no confirmed instance in Section 3 (probability, conditional probability, tree/Venn diagrams) across the facts bank or the thirteen live curriculum files checked here — a genuine absence in the record, not proof it never applies there — nor in Section 1, which the facts bank separately confirms has no standalone question at all. Two failure shapes recur: an intermediate step or table value skipped, or rounded to the target's own precision instead of one place beyond it (Q6(a); Q4(c)); and a reverse-engineered or incomplete result — numbers invented to match a given gap (Q3(b)), or a correct boundary never followed through to a stated count (Q3(c)).

#5. If this topic appears, here's what to expect

Expect a printed target buried inside a larger multi-part question, worth 2–3 of that part's marks, not a standalone question of its own. One habit covers every confirmed variant: never let the working stop at the printed target's own precision, and never let the final line appear without every step that forces it — a standardisation, an interpolation, a named count — shown explicitly on the page. Related but distinct: on an ordinary (non-"show that") part, a bare final answer can legitimately score full marks — confirmed on Jan 2021 Q3(a) ("Ans only 3/3") — but only when it already carries the precision the mark scheme demands ("23.6" vs "awrt 23.60," Q3(b)); that allowance never extends to a part where the target itself is printed in the question.

#Sources and limitations

• Series-level citations trace to research/veridian/WST01-verified-facts.md §5 (Sections 2, 4, 5, 6) and §6; the Jan 2023 Q5(d) citation traces only to paper-anatomy.ts's own audited flag block, not to the facts bank.
• Frequency is stated only as "N of 6 series checked," per this course's standing convention — no percentage or likelihood claim is made anywhere in this document, and no grade-boundary figure is used or implied (none exists in the source material; see signal-frequency-and-boundaries.md §2).
• The 8 series without a located examiner report (Jun 2021, Jan 2022, Oct 2022, Jun 2023, Oct 2023, Jan 2024, Oct 2024, Jan 2025) were not searched for this topic and are excluded from both the numerator and denominator above; Jan 2025 Q1(d)'s "showing your working" instance (mark-scheme-only, no examiner report) is excluded from the frequency count for the same reason.
• This document does not repeat the mechanism derivation, the two full worked exemplars, the four MCQs, or the sub-case taxonomy already in show-that-answer-discipline.ts and codex-deep-dive-show-that-answer-discipline.md; its two new question-level citations (Jun 2024 Q2(b); Jan 2023 Q5(d)) are not present in either.