The Music-and-Bean-Plant Science Fair Project
Ms. Okafor's ninth-grade science fair season is underway, and Priya submits a project titled 'Classical Music Makes Plants Grow Faster.' Priya placed six bean seedlings on the classroom windowsill beside a speaker that played a classical radio station at roughly 70 dB for eight hours a day. Six more seedlings went into a dark storage closet, where the radio could not be heard. All twelve plants were watered with 100 mL every other day and measured after 21 days. Mean height was 18.2 cm on the windowsill and 12.7 cm in the closet, and Priya concludes that music increased growth by 43%. She also ran out of potting mix and used a different brand for the closet plants. In her notebook she wrote, 'This proves music helps plants and my hypothesis is now a theory.' Ms. Okafor wants to respond in a way that honors Priya's effort while teaching how experimental claims are actually built and tested.
1. Identify at least three features of Priya's design that could produce the observed height difference without any effect of music, and explain the mechanism by which each would operate.
Read the worked answer
At least three serious confounds exist, and each is sufficient on its own to produce the 5.5 cm gap:
- Light. The windowsill plants received direct sunlight; the closet plants received essentially none. Light drives photosynthesis, and a closet-grown bean is likely etiolated (pale, spindly, stretched) and light-limited. The 12.7 cm value probably measures darkness more than silence, which makes this the largest problem.
- Soil. A different potting mix means different nutrient content, pH, drainage, and water-holding capacity. A richer mix alone can easily produce a 40% height difference over three weeks.
- Location effects. Closet and windowsill differ in temperature, humidity, air movement, and vibration. The speaker cone also vibrates and warms nearby air, so 'music' is bundled with several physical differences.
- Randomization and replication. Priya did not randomly assign seedlings to positions, used only six per group, and applied the treatment to one room rather than replicating across rooms, so any room-level difference is confounded with music.
- Measurement. A single height reading at day 21 is a weak proxy; dry mass or leaf area would be better, and the person measuring knew which plants were which.
2. Describe how Priya should redesign the investigation so that sound is the only planned difference between groups. Address control, replication, randomization, and measurement choices.
Read the worked answer
A defensible redesign would proceed in this order:
- Common environment. Keep all pots on one bench with the same light, temperature, and pot size, and rotate positions every few days so no plant sits permanently near the window or the heater.
- Same soil and water. Use one potting mix and one pot size, and water by mass (weigh pots and replace transpired water) rather than by fixed volume, so treatments do not differ in moisture.
- Random assignment and replication. Randomly assign plants to treatment and control, with at least 10–15 plants per group. Repeat the whole experiment at least three times (separate cohorts or growth-chamber runs) so the unit of replication is the run, not just the plant.
- A convincing control. The control should receive identical handling without music: a silent speaker in the same room, or a white-noise treatment at matched dB, so 'sound' is separated from 'music.' Measure the actual sound level with a meter at canopy height instead of assuming 70 dB.
- Blind, multi-endpoint measurement. Have someone who does not know the treatment measure height, leaf area, and final dry mass, and report means with variability.
The redesign tests a narrower, falsifiable claim: at a measured sound level, under identical light and soil, does music change growth?
3. Priya's notebook says the results 'prove' her hypothesis and that it is 'now a theory.' Explain what is wrong with both statements and write a scientifically appropriate conclusion for her project.
Read the worked answer
Both statements misuse core terms.
'Proves.' A single experiment cannot prove a hypothesis. Results are always conditional on design, measurement, and assumptions, and alternative explanations (light, soil) remain live. A cleaner experiment would support the hypothesis with a stated confidence level and an effect size, not prove it. Science advances by surviving attempts to falsify, and Priya's design has not yet been tested against a fair control.
'Now a theory.' In everyday speech a theory is a hunch; in science a theory is a broad, well-supported explanatory framework that organizes many lines of evidence (cell theory, plate tectonics, evolution). A hypothesis does not graduate into a theory by being confirmed once. Priya's hypothesis, even if supported, would remain a narrow, local claim.
An appropriate conclusion would be: 'Under these conditions, the windowsill plants grew taller on average (18.2 cm vs. 12.7 cm). Because light and soil also differed, I cannot attribute the difference to music, and I will run a controlled follow-up.' That sentence is honest about uncertainty and points to the next experiment.