# Teacher copy · answer evidence and next moves

Keep this out of learner print packs. [Check A and B](../STUDENT-CHECKS.md) are fresh transfers. Read a source once verbatim if required, but record **self-read, adult-read, prompted, mode and child wording** separately. A pupil's thoughtful answer after listening is valid comprehension evidence and **not yet observed** independent decoding evidence. A missing response may mean access/time failed; re-offer the task with the barrier removed before making a learning inference. Never upload named child work to the public library.

## Day 15 · Check A

| Prompt | Acceptable evidence | Common drift and immediate next move |
| --- | --- | --- |
| 1. Section/structure | **Steps**; numbered/sequence words `First`, `Next`, action verbs `fold`, `write`, conditional `If` are two possible clues. A learner might identify the heading and an action verb. | If “Drawing report” is chosen because it mentions the box, contrast *describes* with *tells what to do*. Read two short sentences and sort them physically; retest on a new two-line example. |
| 2. Navigation | `Questions` heading or the **Box details** menu's questions item; both locate the source limit. Breadcrumb shows page location, not specifically the answer section. No blue-line link is provided. | If learner says blue line, ask whether the line has a destination label/action. Put a paper marker on the heading and move from menu to section; then let child design an informative link label. |
| 3. Supported/unsupported | Supported: “The drawing labels two trays/one handle/a RETURN pocket.” Unsupported: “A real box is light/easy for everyone to carry.” Missing measured weight and willing-user trial. A child may say “The drawn line goes from handle to pocket,” clearly as drawing detail. | If a paper count becomes a real inventory claim, place `the drawing shows` before it. Ask which additional record/test would be needed. If child calls all questions “wrong,” explain questions reveal what needs checking. |
| 4. Word parts | `relabelled` = `re-` + `label` + `-led`: labelled again; double **l** in Australian/British spelling. `unmarked` = `un-` + `mark` + `-ed`: not marked. Accept meaningful equivalent and partial part identification. Record whether whole words were self-read or read aloud. | If prefix reversals are confused, use taught `visible/invisible` then new `marked/unmarked` picture labels. If spelling is the issue, practise `label → labelled` in a short explicit session; do not turn this check into a penalty for a novel pattern not yet taught. |

**Quick mark for planning, not a grade:** For each of `structure`, `navigation`, `source boundary`, `word-part meaning`, use **2** = independent accurate evidence with a reason, **1** = accurate with a prompt/partial reason, **0** = not yet observed after an accessible chance. Write the actual response beside the numeral. Record `reading route` separately. If structure 0–1, begin Day 16 with two headings and their purposes; if source boundary 0–1, return to Card A with a model/real sorting card; if navigation 0–1, give a tangible paper menu/heading path. For word-part 0–1, inspect taught prerequisites and supply a correctly sequenced spelling/reading mini-lesson.

## Day 20 · Check B

| Prompt | Acceptable evidence | Common drift and next move |
| --- | --- | --- |
| 1. Summary | “The paper model is a three-panel exhibition. The front marks an entrance and two picture cards; the back marks a question desk. It has no measurements or visitor observations, so it cannot prove real access.” Any two accurate details plus an explicit limit fit. A labelled AAC/diagram sequence with child wording is equivalent. | If summary imports “inviting” or “easy to use” as fact, highlight the designer's **In my view / If / might**. Rebuild `topic → two report details → limit` on another practice model. |
| 2. View/prediction/question | View: “front panel looks inviting.” Prediction/condition: “If the real corridor were wide enough, some visitors might find the desk easily.” Need corridor width and a trial with willing users. Useful question: “How wide would it need to be?” or “Could a user locate the desk?” | If child treats `might` as an outcome, contrast a possible claim with an observation. If question seeks private visitor information, model consent and nonpersonal measurement first. |
| 3. Framing and limit | Side-view makes **picture cards above a drawn table** easier to see. Neither view establishes actual height/reach, width, lighting or visitor experience; there is no scale/test. | If child says side-view proves reach, ask for a numerical height and an actual user/measure. Distinguish where a symbol is placed from how a person can use it. |
| 4. Word parts | `repositioned` = `re-` + `position` + `-ed`: placed in a different position or positioned again. `unmeasured` = `un-` + `measure` + `-d`: not measured. Accept a sound explanation even if a learner has not mastered spelling of the whole word; record help and taught-pattern status. | If word reading is not yet observed, run an individual, matched text/word probe at a separate time. If meaning is confused, move a paper symbol, describe the new position, then name the action. |

**Planning rubric:** Score `source-bound summary`, `view/prediction and question`, `frame/limit reasoning`, `word-part meaning` 0/1/2 with the same definitions. Preserve the mode and exact language. A child might have strong ideas yet need paragraph transcription support; provide that without changing the idea. If summary is 0–1, explicitly model a new unrelated topic/detail/limit example, then fade supports. If view/prediction is 0–1, sort `drawing fact / personal view / conditional possibility` cards. If image reasoning is 0–1, compare two tactile/printed views of one fresh paper object. If morphology is 0–1, use a short prerequisite check and targeted reading/spelling sequence. Give a child a chance to revise later; store both versions under school policy.

## Check the assessment itself

The two checks sample a narrow slice of English across 25-minute blocks. They do **not** certify the full Year 4 achievement standard, independent text fluency, digital accessibility, multilingual proficiency or disability status. Handwriting, accent, sensory tolerance, speed and family device access are not the targets. A teacher should gather further evidence in school-approved texts and modes, then use the applicable local reporting rules.

**Original resource rights:** © NeuroForgeIO Pty Ltd 2026, SubjectNest, [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Credit, link and indicate changes. Keep this teacher copy separate from student checks.