The Fertilizer Fair Test
A student claims a new liquid fertilizer makes bean plants grow taller. She sets up four pots: Pot 1 gets 50 mL of fertilizer weekly, Pot 2 gets 25 mL, Pot 3 gets 10 mL, and Pot 4 gets plain water. She places Pots 1 and 2 on a sunny windowsill and Pots 3 and 4 on a shaded shelf because that is where space was available. All pots use the same soil and seed type. After four weeks she reports that Pot 1 grew tallest and concludes 'more fertilizer means more growth.' A classmate says the experiment proves nothing. The teacher asks the class to separate what the data can support from what the design cannot.
1. Identify the confounding variable and explain how it could produce the observed result without any fertilizer effect.
Read the worked answer
Confounder: light exposure. Pots 1 and 2 (the higher fertilizer doses) were on the sunny windowsill; Pots 3 and 4 (low dose and control) were shaded. Light is a major driver of plant growth, so the two high-dose pots had both more fertilizer and more light.
Alternative explanation: The sunny pots grew tallest simply because they received more light, regardless of fertilizer. The design cannot separate the fertilizer effect from the light effect because they change together.
What this means: The conclusion 'more fertilizer means more growth' is not supported. The result is consistent with a light effect, a fertilizer effect, or both - the experiment cannot tell them apart.
2. Redesign the experiment as a fair test and state what comparison would actually test the claim.
Read the worked answer
Redesign: Keep all pots in the same light conditions (same windowsill or same grow-light shelf), same soil, same seed type, same watering schedule, same pot size. Vary only the fertilizer amount: 50 mL, 25 mL, 10 mL, and 0 mL (water control). Ideally include multiple plants per treatment to average out individual variation.
The valid comparison: Compare growth across fertilizer levels within the same light condition. If Pot 1 (50 mL) still outgrows the control when light is equalized, the fertilizer claim gains support.
Why it matters: A fair test changes one variable at a time. The original design changed two variables at once, so no comparison in it isolates fertilizer.
3. The student wants to report 'Pot 1 grew 12 cm taller than Pot 4.' What additional information would make that comparison more trustworthy?
Read the worked answer
Needed information:
- Replication: Was 12 cm a single plant or an average of several? One plant per treatment is anecdotal.
- Variability: Report the spread (range or standard deviation) across plants in each treatment. A 12 cm gap with wide spread may not be meaningful.
- Equal conditions confirmed: State that light, water, soil, and pot size were identical.
- Measurement method: How and when height was measured, and whether the same person/technique was used.
Reasoning: A difference between two single plants can reflect individual variation, not treatment. Replication plus a measure of spread lets the class judge whether 12 cm exceeds normal plant-to-plant noise.