Hold up one clear jar of mixed sand, salt and water so the whole class can see the cloudy swirl. Stand a second jar of the same mixture beside it and name it aloud as the baseline (poured through nothing). You will compare later results against it. Leave both jars visible.

Keep this beat light: one question and a quick hands-up of ideas. Do not name variables or hand out kit yet.

Key questions: What can you see in the jar? Which part looks easiest to get out first? What would “fair” mean if two groups tried different ways?

Nature of STEM: Separating mixtures is everyday chemistry, the same idea behind cleaning water and recovering useful materials.

Remind the class that a 6th-class pupil can already turn a theme into a testable question. You set the theme; each group designs the specifics.

This step is oral only: groups choose two methods and say a rough prediction with a because. Do not hand the Investigation Journal page yet. Writing the full question and plan comes in the next step.

Put the model question on the board and leave it up: Does filtering leave less sand in the water than sieving?

Methods on offer (say the plain meaning as you point to the kit):

• Sieving — pour through a sieve or colander
• Filtering — pour through filter paper (or kitchen roll) in a funnel
• Settling — leave the mixture still, then gently pour off the clearer water

Evaporation is not a pupil method today.

Predictions are never wrong. They are the start of the science. Push for a because: “We think filtering will win because the holes are smaller.”

Differentiation: less confident groups pick from two named methods you shortlist; confident groups phrase their own question fully when they write it next (“Does filtering leave less sand in the water than settling when we use the same amount of mixture?”).

Before the Lesson

• Print the Investigation Journal pages, one set per pupil
• Mix one sand–salt–water jar per group plus two class baseline jars
• Set out sieves, funnels, filter paper, clear cups, jugs and timers on a side bench (groups will claim only the tools for their two methods)
• Open the fair-test-planner interactive on the IWB
• Place a class waste bucket for sandy leftover mixture (not for sinks)

Investigation Journal page: required sections

The printable page (or an improvised equivalent grid if needed) must give pupils labelled space for all of the following so you can check the look-fors:

• Our question — testable question naming the two methods
• Our prediction — which method will leave less sand, with a because
• We will change — the separation method only
• We will measure — sand left as cloudiness score 1–5 against the baseline
• We will keep the same — amount of mixture, time rule, container
• Results table — columns for Method, Trial 1 score, Trial 2 score (room for two methods × two trials)
• What we saw (observation) — scores / plain descriptions only
• What we think it means (inference) — kept separate from the scores

Key Concepts

Driving the interactive

Lead with the five cards and the Change / Measure / Keep-the-same sort first. Hang only fair test on that beat before groups write.

Cards on screen: the separation method, how much sand is left, amount of mixture, time allowed, the container.

Target placement: drag the separation method to Change, how much sand is left to Measure, and the other three to Keep the same.

The on-screen prediction line is a neutral stem (Method A vs Method B). Your oral worked model can still use filtering versus sieving.

Cloudiness scale and observation vs inference

Point again at the baseline jar named in Getting Started. Agree the shared cloudiness scale on the board before groups finish their planner page: 1 = clear, 5 = very cloudy (as cloudy as the baseline). Then one short line only: score 2 is what you saw (observation); “filtering won” is what you think (inference). Do not expand this into a second mini-lesson here.

Worked model (say aloud before groups write)

I wonder which leaves less sand, filtering or sieving. I predict filtering, because the paper holes are finer. I will change only the method. I will keep one scoop of mixture, two minutes, and the same clear cup. I will measure cloudiness from 1 to 5 against the baseline jar. After one quick demo pour through filter paper: I observed score 2. I think the paper held back more sand than leaving the baseline alone.

Protect planner time: keep the whole-class sort and scale agreement brisk (a few minutes) so every group leaves this step with methods, controls and the measure rule written before kit is touched.

Checkpoint before kit: do not dismiss any group to the side bench until you have seen, on their Investigation Journal page, (1) two named methods, (2) amount, time rule and container listed as kept the same, and (3) the cloudiness 1–5 measure rule. Less confident groups get a 30-second stem prompt; no group starts pouring without a complete plan.

Kit claim (end of this step / start of next): each group collects only the tools for their two chosen methods (sieve and/or funnel with filter paper; settling groups need clear cups and the timer, not a sieve they will not use).

Misconception to head off: “We should use more mixture for the method we like.” That makes the test unfair. Another: writing “filtering is better” in the results column. That is an inference, not the reading.

Groups also decide whether they will pour off settled water or only judge after a fixed still time. That detail is theirs, as long as both of their methods use the same rule and the still time fits inside the practical window (about two minutes per trial is realistic).

Setup and kit claim

One station per group: mixture jar, two clear result cups, shared baseline jar visible at the front (or one baseline per pair of groups), shared timer or stopwatch. After the planner checkpoint, groups collect only the tools for their two chosen methods from the side bench (sieve and/or funnel with filter paper). Settling-only pairs do not need a sieve sitting unused on the desk.

Pacing (19 minutes is one continuous practical block)

Scores are written during trials, not after. Aim for roughly: 2 minutes kit claim and stir; then four short trials with scoring as you go. For settling comparisons, still-time burns the clock. Cap still time at about two minutes per trial unless both of that group’s methods can share a slightly longer time and you will still finish inside the window. Keep pour-through groups moving so the class is not waiting on one slow station.

Scoring rule (say once, leave on the board)

Hold each result cup beside the baseline jar. Score 1 if the water looks clear, up to 5 if it looks as cloudy as the baseline (or very cloudy). Use that same rule for every trial and every method.

Procedure cues (groups follow their own plan)

• Stir the mixture gently so every scoop is similar
• Use the same scoop or measure for every trial
• Start the timer together; stop at the agreed time rule
• Score cloudiness 1–5 against the baseline before stirring the result
• Write each score on the Investigation Journal page as you go
• Repeat each method once more (two trials each)

What to expect

Filtering usually gives the clearest water (lowest score). Sieving often lets fine sand through. Settling improves with time but two minutes may still look cloudy compared with a good filter. If every score comes out the same, the mixture scoop is too small or the time is too short. Lengthen the still time slightly for settling groups only if both of their methods can share the new time and you still finish inside the practical window.

Safety while circulating

Wipe spills at once. No tasting. Sand stays out of eyes; wash hands at the end. No heat sources for pupils. Pour sandy waste into the class waste bucket, not down the sink.

Fold watchers in: when one group is pouring at the front bench, ask the class: “Do you think their score will beat the baseline? What must they keep the same?”

Fast finishers: sketch one change that would make their test unfair, or say where the dissolved salt still is after the sand is removed (still dissolved in the water).

Recording is required in this lesson, not homework. Most scores should already be on the page from the practical; use this beat to finish gaps and separate observations from inferences before the class discussion.

Circulate and check against the page sections: every group has two trials per method in the results table; scores are clear in the observation space; no one has written “filtering is better” in a results cell (that belongs under inference).

Look-fors:

• Same amount, time rule and container noted as kept the same
• Two readings per method
• Observation and inference separated

If a group’s two trials disagree a lot, ask them what might have changed (scoop size, stirring, timing) rather than calling the test a failure. That is useful fair-test talk for the next lesson in this project.