Four Cups on a Windowsill: What the Bean Data Actually Show

What the data support

• Light matters for color and form: A and C differ only in light, and C is yellow and spindly. That comparison is clean.
• The seed did not need soil to sprout and grow green: B (cotton, no soil) grew green and 9 cm. So 'plants get their food from soil' is contradicted by the class's own Group B.

What the data do not support

• 'The dark plant grew tallest, so light stunts growth.' Height alone is a poor proxy for health. C is etiolated - stretched, pale, and weak - because it is spending stored reserves reaching for light. Taller is a symptom of the problem, not evidence that light is harmful. The students measured the right thing and interpreted it backwards.
• Any claim about soil versus no soil is confounded: D (gravel) died, but gravel holds almost no water, so D may simply have dried out. Soil, cotton, and gravel also differ in water retention, not just in 'nutrients.'

Design flaws

1. No replication. One seed per condition. If a seed was inviable, the group's whole result is noise. Three or four seeds per cup would fix this.
2. Confounded variable in D. Water retention and root support were not controlled.
3. Measurement limited to height. No record of leaf color, leaf number, stem thickness, or mass - the measures that would reveal that C is unhealthy.
4. No structured observation log, so the conclusion rests on a three-week memory.

The conclusion on the chart is therefore not just imprecise; it teaches a misconception the unit test will penalize.

I have two weeks before the unit test. I would not restart from scratch; I would run a revision cycle on the existing cups plus one added condition, which itself models how science works.

Fixes to make now

• Ask each group to plant three seeds in its cup (or add two cups) so we can talk about whether one result is trustworthy.
• Change D to sand or perlite and, more important, have students check and record that all cups receive the same amount and that none is bone dry - or measure water loss by weighing cups. If we cannot control water retention, we say so out loud and label that comparison 'unfair.'
• Add an 'added minerals' condition: cotton plus a dilute fertilizer solution. That directly tests whether the soil's contribution is nutrition rather than 'food,' and it keeps the original question alive.

Recording sheet (daily, 30 seconds) | Date | Height (cm) | Number of leaves | Leaf color (1–4 scale) | Stem: stiff or bending | Notes |

Leaf number and color are the measures that expose Group C. A plant can gain height while losing leaves and chlorophyll.

Then a graphing task that forces the interpretation: plot height and leaf number on the same days. Students will see C's height climb while its leaves yellow and drop, which makes 'tallest means healthiest' visibly untenable. That is a far stronger corrective than being told they are wrong.

That student has made a genuinely good move - he noticed his own data conflict with the class claim and tried to save the claim by weakening it. I would treat that as the engine of the lesson, not an interruption.

In the moment

1. Name the reasoning. 'You just did something scientists do: you found a fact that doesn't fit and you tried to adjust the idea. Let's test the adjustment.'
2. Put the question back with evidence. 'If soil were food, where would the cotton plant's food have come from? It had water, air, light, and a seed. Which of those could be food?'
3. Surface the hidden assumption. Most fourth graders assume, from everyday language, that food is something taken in from outside. Ask: 'What do you eat? What does a plant take in? Is there anything a plant makes for itself?'
4. Do not confirm or deny yet. Record both ideas on a chart: Idea 1: soil is food. Idea 2: plants make their own food; soil supplies water and minerals. Say we will look for evidence.

Where the lesson must land By the test, students should be able to say: a seed contains stored energy that lets a plant start without soil; once leaves form, the plant uses light to make its own sugar from water and air; soil supplies water and mineral nutrients, which is why a plant in cotton can look fine briefly but not thrive long-term. That last point rescues D from being a pointless dead plant - gravel and cotton fail for lack of minerals and water retention over time, not because soil is edible.

The honest caveat to state: 'food' and 'nutrients' are different words, and the everyday meaning of food is what tripped the class up. Naming that confusion explicitly is the instructional move, not avoiding it.