Statistics 1

Question Selector Guide

#WST01 (Statistics 1) — Question Selector Guide

Provenance: written 2026-09-29, built on research/veridian/WST01-verified-facts.md (compiled 2026-09-12, spot-check addendum 2026-09-25), cross-checked against paper-anatomy.ts (content/veridian/curriculum/wst01/), and the existing technique-hub documents in this folder — forge-exam-technique-guide.md, forge-battle-card.md, and signal-frequency-and-boundaries.md — specifically to avoid restating their content. No new PDFs, web fetches, or external searches were performed for this document; every claim below traces to material already verified in those sources. This document does not exist to tell you which questions to skip — see §1 for why that framing is wrong for this specific paper — but to give a real, sourced decision procedure for order of attempt and for what to do when a specific, examiner-documented sticking point costs you time.

#1. Why this is a triage guide, not a "which question to skip" guide

WST01 gives you no question choice. The spec states this directly (§S1.2, "Students must answer all questions"), and the real front page repeats it under Advice: "Try to answer every question." So "selector," on this paper, cannot mean picking a subset — it means deciding what order to attempt the compulsory questions in, and what to do the moment you feel a specific question running long. That is a real, separate skill from knowing the content, and it is not covered by forge-exam-technique- guide.md's own §1 (the flat and per-question timing budget) or its §6 exam-day protocol (which states the general rule "budget against each question's own printed marks" but does not walk through a decision procedure for what happens when that budget is blown). This document is that decision procedure, built from where the real record shows time actually gets lost, not from generic exam advice.

#2. The real time anchor — restated in one line, not re-derived

90 minutes, 75 marks, no Section A/B/C, 6–8 compulsory questions depending on series (confirmed on all 14 front pages reviewed, Jun 2019 → Jan 2025). The flat average is 72 seconds per mark; the one fully mark-broken-down series this course has verified end to end, January 2023, shows the real spread behind that average: Q1 10, Q2 10, Q3 11, Q4 13, Q5 17, Q6 14 marks (75 total). Pearson's own front page states the pacing rule directly: "use this as a guide as to how much time to spend on each question." Full derivation and the complete table: forge-exam-technique-guide.md §1. What follows here does not repeat that budget — it tells you what to do with it once the paper is actually open in front of you.

#3. Rule 1 — identify each question by its content, not by its number

The single most useful, and most consistently overlooked, fact for triage: the same spec topic does not sit at the same question number series to series. This is checkable directly, and the spread is wide enough that memorising "Q5 is always the Normal distribution question" or similar would actively mislead you:

Correlation and regression (spec 4.1–4.3): Q2 in Oct 2021 and in Jun 2022, Q4 in Jun 2024, but Q6 — the very last question — in Jan 2023. Same topic, opposite ends of the same paper.
Probability (conditional probability, tree/Venn diagrams, spec 3.1–3.4): Q1 in Jan 2021 and in Oct 2021, but Q2 in Jan 2023 and split across Q3/Q4 in Jun 2022.
Discrete random variables (spec 5.1–5.4): Q4 in Jan 2021, Q5 in Jun 2022 and in Jun 2019 (14 marks), inferred at Q3/Q4 in Jan 2023 — but the one clean example from a mark-scheme-only citation, Jan 2025, places it at Q1, the paper's opening question, in a 7-question series.

Practical rule: read every question's opening line before you commit to an attempt order. Don't assume the paper is organised the way last year's was. This is not this document inventing caution for its own sake — it is the direct, checkable consequence of the position data above.

#4. Rule 2 — one topic IS positionally consistent: reserve real time for it, wherever it lands

Against that backdrop of genuine variability, one pattern is real and worth trusting: across the 5 series with a fully-read examiner report (Jan 2021, Oct 2021, Jun 2022, Jan 2023, Jun 2024), the Normal-distribution / conditional-probability-under-a-Normal-model content (spec 6.1) has never been seen at Q1 or Q2 — it appeared at Q3 in 1 of the 5 (Jan 2021), and at Q5 or Q6 — one of the last two questions on the paper — in 4 of the 5 (Oct 2021 Q6, Jun 2022 Q6, Jan 2023 Q5, Jun 2024 Q5). It is also, wherever a full mark breakdown exists, the single heaviest question on the paper: 17 of 75 marks in Jan 2023, close to a quarter of the entire exam in one question.

That combination — reliably late, reliably the heaviest single question, and (§5 below) the question every examiner report treats as the most challenging — is the one fact this guide considers load- bearing for pacing. Concretely: do not let an earlier question run over budget on the assumption that "there's less left to do later." By the time you reach whichever question turns out to be the Normal-distribution one, it is disproportionately likely to still be in front of you, disproportionately likely to be worth more marks than the one before it, and — per §5 — disproportionately likely to be where you get stuck. Protecting its time budget matters more on this paper than protecting any other single question's.

#5. Rule 3 — know your bail-out points before you reach them

A triage guide's real job is telling you what to do the moment a question is clearly running long, not just where the time pressure will come from. The record gives three specific, named sticking points worth deciding on in advance, rather than discovering mid-exam:

The multi-step inverse-Normal problem. Jun 2024 Q5(c) asked students to find an unknown parameter from a given probability across several steps; Pearson's own examiner report states plainly that this was "the sticking point for many students" and that "it was not unusual for students to give up at this point." Decision rule: if you recognise this question type (working backward from a probability to a parameter, several steps deep) and you stall, write down the standardisation equation and a labelled sketch before moving on — those earn real M and B marks on their own — rather than leaving the space blank while you keep trying to force the algebra.
The final, most-discriminating part of the paper. Pearson's own examiner report on Jun 2022 describes that series' Q6(c) — a conditional-probability question built on a double inequality under a Normal model — as "the final part of the paper... also the most discriminating part," and records the specific wrong algebra several cohorts produced (treating the double inequality as a straight subtraction the wrong way round). Decision rule: on any Normal-distribution part phrased as a probability between two values conditioned on a third, sketch the curve and shade the regions before attempting any algebra — this is Pearson's own repeated remedy for exactly this error type (§6 below), and it costs seconds, not minutes.
The "given that" conditional-probability trap, specifically under a Normal model. Jan 2023 Q5(e) records students finding the conditioning event's own probability, P(L≤5), and stopping there rather than dividing by it to complete the conditional probability. Jan 2021 Q3(c) shows the related but distinct failure one step earlier — not recognising a conditional probability was needed at all, producing a wrong ratio of the two given raw values instead. Decision rule: the moment a Normal- distribution part contains "given that," write the division by the conditioning probability as your very first line, before you've finished working out either value, so the conditional structure of the answer is on the page even if you run out of time finishing the arithmetic.

None of these three is this document's own invention — each is a specific, named, quoted examiner observation about where real candidates on real series actually lost time or gave up, not a generic "read the question carefully" platitude.

#6. Rule 4 — the safest opening move the record actually supports

Data representation and summary (spec 2.1–2.4 — histograms, stem-and-leaf, mean/SD, quartiles, outliers, box plots, comparing distributions) is the only spec topic confirmed present in every one of the 6 examiner-report series this course has read (Jun 2019, Jan 2021, Oct 2021, Jun 2022, Jan 2023, Jun 2024), and it is the smallest question wherever a full breakdown exists (10 of 75 marks, Jan 2023). Narrowed further to its most commonly tested sub-skill — comparing two distributions in words — this course's own signal-topic-brief-reading-and-comparing-distributions.md confirms it sits at Q1, the paper's opening question, in 3 of the 4 series that test it (Jan 2023, Jun 2022, Jun 2024); the one exception, Oct 2021, places the equivalent content at Q3.

Put together: if the question in front of you is asking you to read a histogram, stem-and-leaf diagram, or box plot, or to compute a mean/SD/IQR from one, it is reasonable to treat it as the paper's opening question and a good place to bank secure, fast marks before the paper's heavier, more variably-positioned content arrives. Flagged plainly: this specific "open with it" recommendation is this document's own inference from the frequency-and-position data above, not a sentence any examiner report states outright — the examiner reports describe what goes wrong within Section 2 questions (histogram-area errors, stem-and-leaf reversal, wrong outlier formula — see forge-battle-card.md §4 for the full list), not where a candidate should choose to start.

#7. Rule 5 — a "show that" item under time pressure is a zero-or-full-credit gamble, not a partial one

This matters specifically for triage because the instinct under time pressure is to shortcut exactly the kind of question where shortcutting costs the most. The M/A dependency mechanism documented in paper-anatomy.ts (an A mark cannot be earned without its own M mark already secured — "M0 A1 is impossible") bites hardest on a "show that" item, where the target value is printed in the question: Jan 2023 Q5(d) is coded M1, A1ft, A1*, and the starred final mark is annotated as needing a fully correct solution with no errors. Oct 2021 Q6(a)'s examiner report records the real failure mode this produces under pressure: many students wrote the standardisation step, then jumped straight to the final value, "missing out the intermediate step and the accurate answer" — and some wrote only the final decimal with no working at all. Decision rule specific to triage: if you are short on time on a "show that" part, writing the one missing intermediate line is a better use of remaining seconds than polishing a part you've already fully answered — the intermediate step is where the M mark, and therefore every A mark chained to it, is actually won or lost; a bare final answer that matches the printed target can still score zero.

#8. Rule 6 — the one accuracy pass that costs no extra time

Not a triage decision so much as a standing instruction worth running on every question regardless of how much time is left, because it costs nothing: quote table values to 4 decimal places and round final inexact answers to 3 significant figures unless told otherwise — the front page's own rounding instruction, and, per the verified-facts bank, the single most-repeated accuracy trap across every Normal-distribution examiner report reviewed (Jan 2021 Q3(b), Oct 2021 Q6(b), Jun 2024 Q5(b), each independently flagging students substituting a rounded z-value like 1.64 for the table's 1.6449). This belongs in a selector/triage guide only because it is the rare piece of exam technique that has zero time cost against the budget in §2 — there is no trade-off to weigh, so it should never be sacrificed even on a question you are rushing.

#9. What this guide does not, and cannot, tell you

No fixed topic-to-question-number map exists. §3 above is the evidence for why one specific fixed map would be actively wrong to teach — the position data is real, but it describes a range, not a rule you can apply to a paper you haven't opened yet.
Examiner-report coverage is 6 of 14 series, not all of them (Jun 2019, Jan 2021, Oct 2021, Jun 2022, Jan 2023, Jun 2024 read in full; the remaining 8 searched for but not located as of this revision — see WST01-verified-facts.md §3/§8). Every "seen in N of M" figure in this document uses that real denominator, never all 14, and never a percentage.
No data exists anywhere in the source record on candidates' actual minute-by-minute pacing — the sticking-point evidence in §5 comes from examiner reports describing what went wrong and, in one case, that students visibly gave up, not from any timed-pacing study. This document infers time-pressure triage rules from content difficulty and question position; it does not have, and does not claim to have, direct timing-under-exam-conditions data.
No grade-boundary or likelihood-percentage data exists in the source material, and none is invented here — consistent with signal-frequency-and-boundaries.md §2's own finding that no grade-boundary figures exist anywhere in this course's WST01 research.

#Sources

research/veridian/WST01-verified-facts.md §2 (assessment structure), §5 (per-topic verified content, all series/question citations used in §3–§8 above), §6 ("show that" cross-cutting pattern), §7–§8 (topic strength and flagged gaps); content/veridian/curriculum/wst01/paper-anatomy.ts (Jan 2023 per-question mark breakdown, M/A/B mechanism, Q5(d) A1* example); content/veridian/technique-hub/ wst01/signal-frequency-and-boundaries.md (the 6-series examiner-report denominator and per-topic mark-weight evidence); content/veridian/technique-hub/wst01/signal-topic-brief-reading-and- comparing-distributions.md (Q1-vs-Q3 position data for the compare-in-words sub-skill); content/veridian/technique-hub/wst01/forge-exam-technique-guide.md §1 and §6 (the timing budget and general exam-day protocol this document deliberately does not restate); content/veridian/technique- hub/wst01/forge-battle-card.md §4–§5 (cross-referenced, not repeated, for the full common-error list and topic-weight summary).