# Ten flexible teacher starts · Days 21–30

Read [the small setup](MATERIALS.md) once; select a currently useful [participation route](PARTICIPATION-ROUTES.md). Each start suggests 25 minutes and can be split, repeated, shortened or stopped. No finish/speed score. Reuse familiar approved apparatus only when available/wanted. Each ordinary lesson offers a short actual shadow opportunity before diagram discussion; an exact source/model/plan route remains complete for its stated evidence. Do not replace absent observations with ideal values. Learner invitations are in [LEARNER-CARDS](LEARNER-CARDS.md).

## Day 21 · A shadow for my idea

**Job:** choose a shadow-picture/message or a question; notice one actual effect. **Ready:** approved gentle light, broad opaque card and pale screen, or light-parts description. **Say:** “What could our shadow do?”

- **0–3:** Offer shadow-picture, a pretend doorway/message, inquiry alone or an own idea. Agree whether actual light is wanted; an adult can do all positioning. Stop remains available.
- **3–8:** Make one brief actual shadow encounter in the approved setup. Invite “What do you notice?” Accept unclear/soft/dark/changed or no available observation. Ask the learner which part interests them; do not supply a measurement or pretend they operated equipment. Without apparatus, encounter the source description and choose an own question instead.
- **8–12:** Name the light source, opaque blocker and receiving screen in this setup. Explain that blocked light can make a darker area; other light can still reach it. Introduce the ideal drawing only after the actual/source encounter, with its model label.
- **12–20:** Follow one chosen idea: a broad shadow shape, a doorway-like gap between safely positioned broad cards, or a question about size/position. Learner observation/direction can be the meaningful work. Try only locally approved changes; a proposed change can stay a plan.
- **20–25:** Ask “Keep, change or stop?” Notice one actual choice and one available observation/source response. Next invitation: ask which part makes the light. A nice supplied picture is not an observed learner result.

## Day 22 · Which part gives the light?

**Job:** connect an actual source/blocker/screen to the chosen idea. **Ready:** same approved setup or light-parts exact description. **Say:** “Where does the light start?”

- **0–3:** Restore only a wanted available idea; otherwise choose a fresh imaginary window/shape. Identify actual roles and comfortable source access.
- **3–8:** Offer another short actual shadow look before labels. Let the learner identify the emitting source in the actual setup; a pale receiving screen or mirror can reflect light without emitting it. A glowing display may itself emit light; identify the actual object and source. If switching is safe and useful, an adult may briefly change only the lamp state while preserving comfortable room light. Switching is optional, never a required child action.
- **8–12:** Use the light-parts aid to label this source, blocker and screen. An unlit torch drawing alone does not tell us whether the real torch was on. Name other available light when it affects the view; keep observation and explanation distinct.
- **12–20:** The learner chooses a clearer position/view or a different suitable receiving surface for their shadow job. Predict what that might help show, then optionally try. If two things change, name both; this exploratory try need not be called a fair comparison.
- **20–25:** Invite one source explanation in the learner's usual mode. Record the actual choice/reason/help; keep “source unclear” where appropriate. Next: make a model of the path, not a copied actual measurement.

## Day 23 · An idea about the path

**Job:** model straight light paths and ask what the model leaves out. **Ready:** ordinary Shadow Window aid; actual apparatus optional. **Say:** “What might reach the screen?”

- **0–3:** Choose one part of the learner's shadow job to think about. Recall actual apparatus versus ideal model with two plain names, “our lamp” and “our drawing.”
- **3–8:** Offer one brief actual shadow observation if approved; let the learner choose which edge/place to notice. Without apparatus, begin with the exact named source and a prediction, retaining no actual observation.
- **8–12:** Encounter the original ordinary Shadow Window source card. It has point source0, bar4 and screen8 in arbitrary model units; show ideal straight rays past the bar edges. The learner may trace, describe or direct a line; no handling, count, ratio or numerical height calculation is required.
- **12–20:** Make an own model of a chosen blocked path, or change/reject the supplied model and say why. Try an actual view again if useful, keeping results separately attributed. Ask one real limitation: “Is our lamp a tiny point? Are these edges clear?” Accept an incomplete explanation and a different next question.
- **20–25:** Notice an actual model choice/reason and available actual view, not an automatic inquiry award. Keep the model's author purpose separate from what it helped this learner do. Next: choose one thing to change in a comparison.

## Day 24 · Change one thing on purpose

**Job:** choose a repeatable blocker-position comparison for the child's purpose. **Ready:** same approved setup or ordinary change-one model; positions chosen locally. **Say:** “What shall we change? What stays?”

- **0–3:** Choose a question about the blocker, such as whether moving it while keeping the same source and screen changes the visible shadow. Accept another investigable question and name a relevant outcome to notice.
- **3–8:** Before diagrams, offer a short actual baseline shadow look. Agree source, screen, blocker size/orientation and what position reference can be kept. No real distance comes from the diagram. Name learner/adult operators.
- **8–12:** Predict a possible effect and reason. If useful, encounter the ordinary change-one idea model (bar4→2, screen8 fixed); it is not held C/D and not actual data. The learner can use a qualitative relation rather than numbers.
- **12–20:** Optionally perform the chosen comparison and repeat a useful part under agreed conditions. An adult can move the blocker; the learner can observe/direct/record. If an edge is unclear, choose a qualitative description or stop. If the lamp/screen also moved, retain that fact and avoid a single-variable conclusion.
- **20–25:** Keep one actual available finding or an unperformed plan. Ask what would help another try. First predictions remain separate from observations, and a changed prediction is welcome. Prepare held C privately for Day25.

## Day 25 · Fresh C: a new question

**Job:** first own response to a new source, then optional further work. **Ready:** only [fresh C brief](STUDENT-CHECKS.md#fresh-c--a-new-shadow-question)/fresh-c aid; teacher key/model data held back. **Say:** “What would you like to find out?”

- **0–3:** State that this is a new ideal point-source drawing, not a tested lamp setup. Read/describe all source facts in the chosen mode. No actual trial or permission is supplied. Note relevant earlier access to the source; do not present practiced work as fresh.
- **3–12:** Invite one own question/change, prediction/reason or plan component. Offer a scribe/point/direct-description route without supplying the comparison or result. Capture the actual first response and help before another component. Own valid questions, alternative sources and rejecting this source are available.
- **12–20:** After first capture, invite a next component or leave it for another short sitting. A local approved actual try can be planned/conducted only under the ordinary setup review, with actual observations/operators separate. Teacher key examples are available only after first capture as teaching, not backfilled first evidence.
- **20–25:** If unfinished, record **not yet observed**, offer another short sitting and give no completion-speed score. Use the separate descriptive key/next move against what actually happened. The learner need not calculate ideal spans, try all changes or complete a product.

## Day 26 · A record worth keeping

**Job:** repeat a useful actual comparison and represent the available evidence. **Ready:** chosen locally approved setup and optional blank record-a-try mat. **Say:** “What did we actually notice?”

- **0–3:** Choose an available question/plan from ordinary work or begin a fresh one. Never invent a Day25 result. Decide whether actual apparatus is wanted and which observation/quantity matters.
- **3–8:** Offer a short actual baseline/changed look before table discussion. Repeat a part if that will answer the question. Name operators/observers; do not call an adult trial independent child equipment work.
- **8–12:** Choose a fair description/measurement for the available edges: qualitative size/clarity/position, or an appropriate instrument/unit and sensible precision. Let the learner decide how to represent it—table, labelled drawing, spoken/signed record or directed marks. Adult-read measurements retain their source.
- **12–20:** Record the available actual observations, repeat only when useful, and retain missing/unclear entries. Add model predictions in a separately named area if wanted; do not fill actual-result rows with them. A diagram-only route represents model information and proposes a future record.
- **20–25:** Ask which record helps the child's job/question. Notice an actual representation choice and supported relation, or an honest gap. Next: a different question about the screen. No minimum number of trials or full table gate.

## Day 27 · A different question: the screen

**Job:** test or plan a new screen-position relationship rather than repeating blocker work. **Ready:** approved setup; own broad position descriptions. **Say:** “What if the screen changes?”

- **0–3:** Choose what to investigate about the receiving screen, or propose another distinct question. Check actual source/blocker positions and room arrangements locally.
- **3–8:** Offer a short actual look at the existing screen before discussing a model. Identify a useful outcome—shadow height, placement or clarity—without promising an exact ideal result.
- **8–12:** Make a prediction and agree to retain source/blocker size/orientation/position for this comparison. Name the screen change and what would be observed. Use an own description/model; the held D before/after facts are not practice data.
- **12–20:** Optionally try/repeat the chosen screen comparison. The learner can decide, observe, use an approved instrument or direct an adult. Record if source/blocker also moved or edges changed. The result may be inconclusive; a plan stays a plan if no try occurred.
- **20–25:** Compare the available result with the original prediction. Invite a reason to keep/revise or another question, not a demand to agree with the ideal drawing. Keep this new-variable evidence distinct from earlier blocker evidence.

## Day 28 · What can our evidence say?

**Job:** compare available methods/findings and choose a useful next question. **Ready:** actual records if available; another appropriate attributed record/method only if it genuinely exists. **Say:** “What do we know? What is still open?”

- **0–3:** Select one question for the child's shadow purpose. Check what actual records exist and who made them; a missing page/trial is not reconstructed from memory as a measured result.
- **3–8:** Offer a short approved actual look/retry to clarify one useful uncertainty before paper comparison. If there is no actual source, name the exact account/model being interpreted.
- **8–12:** Compare two available records/methods, including another investigator's only when available and attributed. Two ideal drawings are models, not two observed experiments. Ask what conditions/edge definitions/units are comparable and what differs.
- **12–20:** The learner chooses a justified conclusion, a tentative relation or insufficient evidence. Consider alignment, ambient light, extended source or changed conditions as possible limitations without asserting a cause not observed. Choose a further question or a simpler comparison. No mandatory full error list or numerical average.
- **20–25:** Keep an actual evidence-led reason and a next question. If no other investigator source was available, leave that I05 component open. Next: apply the available evidence to the learner's own design.

## Day 29 · Make my shadow idea useful

**Job:** make, prefer or revise the child-owned shadow design. **Ready:** wanted available work/setup or a new own idea; optional my-shadow-message aid. **Say:** “What would you like it to show?”

- **0–3:** Choose shadow art, a pretend clue/message, a light question or inquiry alone. Select an intended reader/viewer only if wanted; no public audience required.
- **3–8:** Offer one short approved actual making/observation route before design discussion. Let the learner choose a familiar broad blocker/screen arrangement; an adult can position it. Record actual shadow/art actions, not an invented finished design.
- **8–12:** Use one available observation or model prediction to choose a next change. A promised audience effect is only an intention. No actual observations available? Develop an honest design/communication plan instead.
- **12–20:** Make/revise a meaningful component or rehearse a brief own explanation. If a willing reader is genuinely available and the learner agrees, ask what they actually notice; otherwise keep the intended meaning without fabricated feedback. Keep or reject the revision. Do not require a letter-shaped shadow, neat drawing or achieved clarity.
- **20–25:** Prefer, keep, change or stop. Name one actual learner decision and the evidence/source/help used. A private unfinished idea remains useful. Prepare held D privately for Day30.

## Day 30 · Fresh D: a fairer comparison

**Job:** respond to a different unperformed comparison proposal. **Ready:** only [fresh D brief](STUDENT-CHECKS.md#fresh-d--a-new-comparison-plan)/fresh-d aid; key/model data held back. **Say:** “What could you change in this plan?”

- **0–3:** Read/describe both exact source configurations and the proposed claim. It is a new model proposal, not an earlier learner trial. No prior result or local permission exists in the brief; geometry does not command a real setup.
- **3–12:** Invite one own revision/question/reason. Capture the first response and help before giving examples or inviting the next component. Do not require a particular position, numerical ideal calculation, complete fair-test checklist or exact wording.
- **12–20:** After capture, invite a prediction/evidence need, another component or later short sitting. If wanted, plan a locally reviewed actual comparison; actual conduct/observation remains separately dated and attributed. The key's multiple valid examples support teaching after first capture.
- **20–25:** Unfinished = **not yet observed**; offer another short sitting, no completion-speed score. Identify available reasoning and one appropriate next invitation, with actual trial/equipment and missing parts still open. Ten authored starts do not certify the Science year.