# Teacher copy · Year 6 English answers, evidence notes and next moves

Keep this page separate from the [clean learner page](../LEARNER.md), [choices](../DAILY-CHOICES.md) and [fresh checks](../STUDENT-CHECKS.md). Read-aloud, screen reader, speech, sign/AAC, typing and child-directed adult transcription can all make the same question accessible. Record **reading route** separately from reasoning: `exact first response | source named | access provided | content hint, if any | next move`. A response after listening is not by itself evidence of independent reading. None of these invented sources proves a real-world impact, product safety, local policy or student achievement standard.

## Daily exits · expected reasoning and immediate next move

| Day | Expected evidence | If not yet, teach and recheck |
| ---: | --- | --- |
| 11 | A and B are **not independent**: the same fictional Dockside Fix Club produced both from its own trial; A invites, B records. `11 + 4 + 3 = 18`. | Give a fresh fictional poster/log pair and have the learner trace creator and date before interpreting numbers. |
| 12 | Larger `ELEVEN` changes attention, not `11 + 4 + 3 = 18`; four waiting and three not repaired are still printed smaller. | Read the complete text/layout route in order, then ask what a skimming reader may miss on a new invented page. |
| 13 | B records 11 working at closing; C2 says Ren wanted a quieter explanation; neither establishes everyone's preferences or a one-week outcome. | Sort one fresh statement under `records / suggests / not shown`, naming its exact source. |
| 14 | “All 18 were saved” overreaches at **all/saved**. Safe repair: the club recorded 11 working at closing, four waiting for parts and three not repaired then. | Rebuild the closing categories with three source cards; recheck a new `always/everyone` sentence. |
| 16 | Bounded conditional recommendation such as “If volunteers and a quieter option are available, another small trial could be tested.” No new approval or cost claim. | Give the audience/decision in one line and ask for an `if` condition before evidence. |
| 17 | A clear structure gives reader recommendation, evidence, limitation/counterpoint and next step; the huge-program/Jo sentence makes an unwarranted leap and treats one selected view as representative. | Label four paragraph jobs on a new mini-case, then let learner choose a defensible order. |
| 18 | “The club log **records** four items waiting for parts at closing.” In “Although the poster is bold, the small line remains important,” comma follows the opening dependent clause. | Rehearse one new records-versus-proves sentence and a separate opening-clause comma; score meaning and punctuation separately. |
| 19 | Revision names a source boundary or condition, e.g. **at closing** or **if volunteers are available**. A delivered pitch is noted separately from a cue-card plan. | Ask one neutral question on a new fictional pitch; invite a private second try in chosen mode. |

## Thirty practice routes · acceptable responses

| Route | Checkable response or criterion |
| --- | --- |
| D11-A | A and B share **Dockside Fix Club** and **18 September 2026**; A is public invitation, B internal tally. A may be read by potential visitors, B by organiser/staff. Both use one trial. |
| D11-B | Text alternative names the large **ELEVEN WORKING BY CLOSING TIME!**, smaller four/three line and fictional club/footer; one accurate number. No invented visual colour. |
| D11-C | Same club supplied poster and tally, so the number is repeated, not independently verified. A later independent follow-up could ask whether items stayed working. |
| D12-A | All three counts remain 11, 4 and 3; redesigned placement makes the non-working/waiting categories as findable as the lead count. Still no later durability/cost evidence. |
| D12-B | 18 of 30 zines were taken and 12 remained in Mini-case E; “most loved in town” is not established. A ten-word fair headline is “Eighteen of thirty zines were taken at the trial shelf.” |
| D12-C | Sequence names headline first, `11/18` context next, smaller `4/3` and limit last; explains likely attention without calling the accurate figures false. |
| D13-A | Example: **recorded** 11 working at close (B); **suggested** another small trial could help (D); **not shown** whether repairs lasted a week. Question asks for follow-up or capacity. |
| D13-B | Mini-case G supports `32 of 40 labels readable after one indoor day`; no weatherproof claim. An approved outdoor/longer test would address durability, though no result is known. |
| D13-C | A question about one-week condition follows B's missing follow-up; a question about a quieter explanation follows C2 or D; neither receives an invented answer. |
| D14-A | Caption distinguishes A's invitation and B's tally: 11 working at close, 4 waiting, 3 not repaired; no claim that all seven “failed forever.” |
| D14-B | “Sixteen of 20 returns were logged by closing; four were not yet logged.” Ask whether/when those four were later logged; do not label borrowers or items lost. |
| D14-C | Jo liked asking questions; Ren wanted a quieter explanation. Two selected slips cannot represent all visitors, and selection/visitor total is unknown. |
| D15-A, later | Hall Editors' sign invites another shelf; 18/30 taken at close; what readers loved or finished remains unknown. |
| D15-B, later | Both `16 logged` and `4 not yet logged` visible; the sign avoids saying all were safely home, lost or late. |
| D15-C, later | A longer or relevant weather exposure test would address “weatherproof”; current indoor one-day count is 32 readable and 8 not. No outcome is invented. |
| D16-A | Bounded small-trial plan cites 11/18 at closing, one selected Jo/Ren view with attribution, and an if-condition/unknown about parts, volunteers or access. Waiting for evidence is also defensible if reasoned. |
| D16-B | 18 of 30 zines taken supports **testing** another shelf, not “most loved in town”; later reading/preferences remain unknown. |
| D16-C | 16 of 20 returns logged at close; a reporting improvement can include a later check for four unlogged returns, without blaming anyone. |
| D17-A | Four distinguishable jobs: limited recommendation, B/C linked evidence, what is unknown, and a test/approval question. Wording/order can vary if reader can find each. |
| D17-B | The opening admits no reading-completion data; 18/30 taken is a limited reason for a small next trial. Counterpoint-first order can build trust by naming a limit early. |
| D17-C | Source G records 32/40 readable after one indoor day; next stage proposes relevant approved weather testing, not mass purchase. |
| D18-A | Example: “Although eleven items worked at closing, the log does not show whether they lasted.” **Records** is more precise than proves for a tally; comma follows the opening dependent clause. |
| D18-B | Example: “The four discs that had no logged return at closing need checking.” The restrictive `that ...` clause identifies the discs and has no enclosing comma. Do not write “four lost discs.” |
| D18-C | Example: “The indoor note records 32 of 40 readable after one day, so the labels **might** work longer; rain performance is unknown.” A stronger prediction needs a real test; no claim of guaranteed survival. |
| D19-A | A 30–45-second claim repeats a short proposal intentionally, cites B 11/18 at closing, gives an if-condition/unknown, then revises after a neutral question. Record delivered versus planned. |
| D19-B | Three boxes are claim, 18/30 trial evidence, and unknown/next test. A delivered private pitch may support presentation; a cue card alone is planning. |
| D19-C | Questions ask about outdoor rain/long-term evidence; response stays within one-day indoor 32/40 and changes “weatherproof” to a bounded statement. |
| D20-A, later | Four-part note cites 20 lent, 16 returns logged by close, four not yet logged, unknown later status, and a reversible follow-up/logging step. |
| D20-B, later | Another small shelf trial may be recommended with 18 taken/12 remaining; “everyone loves these” is repaired to a source-limited statement and future question. |
| D20-C, later | B/C/D support a possible **small test**, counterpoint is durability/capacity unknown, and the response proposes follow-up or an approval question, not guaranteed success. |

## Day 15 held-out check · Borrow Box

| Item | Acceptable response | What to observe |
| ---: | --- | --- |
| 1 | The **fictional West Room lending team** made both after its first trial. The public poster addresses potential visitors and invites interest in another day. It and its log are not independent accounts. | Creator/context/purpose before accepting a headline. |
| 2 | **“A KIT FOR EVERYONE!”** is largest at top; **three remained** and **no all-user suitability survey** appear smaller below. Placement may foreground invitation and reduce notice of limits; it cannot change `9 + 3 = 12`. | Feature linked to a possible effect, with the complete text-layout route accepted. |
| 3 | Log supports “Nine of twelve kits were borrowed during the day” and “three remained at closing.” It cannot show that each borrower was satisfied, that all visitors found a suitable kit or that borrowed kits were used. Nine borrowed kits are **not** nine satisfied people. | Accurate narrower statement plus explicit unknown. |
| 4 | Ask who wanted a kit but could not use one, whether borrowed kits were useful, or how many would attend another day; a later anonymous voluntary feedback/source check could address that. Do not invent results. | Question/source directly addresses missing decision evidence. |

## Day 20 held-out check · pop-up reading cart

| Item | Acceptable response | What to observe |
| ---: | --- | --- |
| 1 | **Source X** records 30 set out, **19 taken**, **11 left** at closing. **Source Y** is one selected fictional visitor who said large headings helped *them*. Neither proves actual reading, usefulness for all visitors or future demand. | Exact X detail, attributed one-person Y view and unknown. |
| 2 | A second small trial **if** capacity/access and feedback collection are arranged is defensible. Waiting for more information is also defensible if supported by the unknowns. Source Z's “every room/everyone” is an overreach, not approval. | Bounded decision and relevant condition/question. |
| 3 | Model four-sentence response: “**Although Source X records 19 of 30 guide cards taken at closing,** it does not show whether they were read. **Source Y says one visitor found large headings helpful,** but this selected view cannot represent everyone. I recommend a second small trial **if** staff and accessible formats are available. The group should ask more visitors and record usefulness before considering other rooms.” Other sequences/modes are valid if source, limit, recommendation and next step are checkable. The comma after opening `Although` clause is correct; `records`/`says` fit the sources. | Paragraph/argument stages, source-linked evidence, counterpoint, precise verbs and revision. Do not demand this exact wording or handwriting. |

## Formative rubric and next moves

For Day 15, note **source context**, **feature → meaning**, **supported/unknown** and **next question** separately. For Day 20, note **two source uses**, **bounded recommendation**, **limitation/counterpoint**, **argument organisation** and **language edit** separately. Use `I` for accurate independent reasoning with available access, `P` when a content prompt led to the answer, `R` for a specific mismatch and `N` for not yet seen because access/time was missing. Keep an exact phrase or mark beside the code; do not collapse a writer into one numerical mark.

- **Poster treated as independent proof:** compare creator/footer/date of two texts, then test an unrelated new source pair.
- **Selected comment treated as a poll:** ask how many comments/visitors are actually known and resample with a new two-slip case.
- **Layout treated as a truth test:** keep words/numbers identical while changing size/order, then ask what changed and what did not.
- **Universal claim or invented outcome:** add `at closing`, `according to Source X`, `might`, or an explicit `unknown`; recheck a different fictional case.
- **Argument missing a reader's decision:** ask “What should this reader do next?” before polishing sentences.
- **Sentence boundary/comma:** make one new `Although ... , ...` pair; keep semantic reasoning score separate from punctuation support.
- **No source linked to a number:** return to source ID/line; do not reward a plausible number from memory if not attributed.
- **Access not ready:** offer the full text/layout route, private response, supported reading or scribe later; do not treat an unfinished blank as failure.

For family feedback, name one next reading/writing move, not a permanent “visual learner”, confidence or ability label. Classroom and EAL/D review remain pending. Original staff keys and rubric © NeuroForgeIO Pty Ltd 2026, [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).