Scientific Method Examples

Example 1: Testing Whether Handwashing Prevents Disease

In the 1840s, Hungarian physician Ignaz Semmelweis observed that women giving birth in one clinic at Vienna General Hospital died of childbed fever at rates far higher than women in a second clinic. The first clinic was staffed by medical students who also performed autopsies; the second was staffed by midwives who did not.

Observation: mortality rates differed dramatically between the two clinics. Question: what caused the difference? Hypothesis: medical students were carrying "cadaverous particles" from autopsies to the delivery ward on their hands, causing infection. Experiment: Semmelweis required all medical students to wash their hands in a chlorinated lime solution before examining patients. Result: mortality rates in the first clinic dropped from roughly 10% to under 2%, matching the second clinic. Conclusion: hand contamination was transmitting disease, and handwashing could prevent it.

This example illustrates several important features of the scientific method. Semmelweis started with a careful observation, formed a specific hypothesis that could be tested, designed an intervention, measured the outcome, and drew a conclusion supported by the data. His work also illustrates the social challenges of science: his findings were initially rejected by the medical establishment, which was not ready to accept that doctors themselves were causing harm.

Example 2: Testing Plant Growth with Different Fertilizers

A student wants to know which fertilizer produces the tallest tomato plants. Observation: different fertilizers contain different nutrients in different proportions. Question: does the type of fertilizer affect how tall tomato plants grow in 30 days? Hypothesis: plants given fertilizer A (high nitrogen) will grow taller than plants given fertilizer B (balanced formula) or no fertilizer.

Experiment design: plant 30 identical tomato seedlings in identical pots with the same soil. Randomly assign 10 plants to each group: fertilizer A, fertilizer B, and no fertilizer (control). Give all plants the same amount of water and sunlight. Measure plant height in centimeters every five days for 30 days.

Results: after 30 days, the fertilizer A group averaged 28.3 cm, the fertilizer B group averaged 31.7 cm, and the control group averaged 19.4 cm. Statistical analysis showed both fertilizer groups grew significantly taller than the control, and fertilizer B produced significantly taller plants than fertilizer A.

Conclusion: both fertilizers promoted growth compared to no fertilizer, but the original hypothesis was not supported because fertilizer B outperformed fertilizer A. This result suggests that balanced nutrition may be more important than high nitrogen alone. Further experiments could test individual nutrients to identify which ones drive the observed differences.

Example 3: Fleming and the Discovery of Penicillin

Alexander Fleming's discovery of penicillin in 1928 is often presented as a lucky accident, but it actually demonstrates the scientific method at work. Fleming observed that a mold contaminating one of his bacterial culture plates had killed the bacteria surrounding it. Many researchers would have discarded the contaminated plate, but Fleming recognized the observation as significant.

Observation: bacteria near a mold colony died. Question: does the mold produce a substance that kills bacteria? Hypothesis: the mold secretes an antibacterial compound into the surrounding medium. Experiment: Fleming cultured the mold (Penicillium notatum) and tested its broth against various bacterial species. Result: the broth inhibited growth of many disease-causing bacteria. Conclusion: the mold produced a substance, which Fleming named penicillin, with powerful antibacterial properties.

It took another decade of research by Howard Florey and Ernst Boris Chain to purify penicillin sufficiently for clinical use, demonstrating that the scientific method rarely produces immediate applications. The journey from observation to practical result involved multiple researchers, thousands of experiments, and iterative refinement of methods.

Example 4: The Expanding Universe

In the 1920s, astronomer Edwin Hubble made observations that fundamentally changed our understanding of the cosmos. Using the Hooker Telescope at Mount Wilson Observatory, Hubble measured the distances to several galaxies and the speed at which they were moving relative to Earth. Observation: nearly all galaxies were moving away from Earth, and more distant galaxies were moving away faster. Question: what explains this relationship between distance and speed?

Hypothesis: the universe itself is expanding, causing everything within it to move apart, like dots on a balloon that is being inflated. Prediction: if the universe is expanding, there should be a linear relationship between a galaxy's distance and its recession velocity. This prediction could be tested by measuring more galaxies.

Result: Hubble's data showed a clear linear relationship, now known as Hubble's Law, confirming the prediction. Subsequent measurements by many astronomers over decades have refined and confirmed this relationship. The expanding universe model became the foundation for the Big Bang theory, which in turn predicted the cosmic microwave background radiation, discovered in 1965, providing further confirmation.

Example 5: Everyday Scientific Method

You do not need a laboratory to use the scientific method. Consider a simple everyday scenario: your car will not start on a cold morning. Observation: the engine cranks slowly and fails to start. Question: why will my car not start? Hypothesis 1: the battery is weak from the cold temperature. Test: try jump-starting the car from another vehicle. Result: the car starts. Preliminary conclusion: the battery was the problem.

But a good scientist considers alternative explanations. Hypothesis 2: the starter motor is failing. If this were the case, jump-starting might also help by providing extra current, so the first test does not fully distinguish between hypotheses. Additional test: after the car runs for 30 minutes (recharging the battery), turn it off and try starting again. If it starts normally, the battery was weak but functional when charged, supporting hypothesis 1. If it still cranks slowly, the starter motor may be the problem, supporting hypothesis 2.

This everyday example illustrates the same logical structure as multimillion-dollar research projects: observe a phenomenon, form an explanation, test it, consider alternatives, and refine your understanding. The scientific method is not a special ritual reserved for laboratories; it is organized critical thinking applied to any question about how the world works.

Example 6: Citizen Science and Bird Population Monitoring

The Audubon Society's Christmas Bird Count, running annually since 1900, demonstrates the scientific method applied at continental scale through citizen science. Observation: bird watchers across North America noticed that some species seemed to be declining while others were increasing. Question: how are bird populations actually changing over time across the continent? Hypothesis: specific species are declining due to habitat loss, pesticide use, and climate change, while adaptable species are expanding their ranges.

Method: tens of thousands of volunteers follow standardized protocols to count every bird they observe within designated 15-mile-diameter circles during a specific two-week period each December. The standardized protocol ensures data comparability across locations and years. Data analysis reveals long-term population trends for hundreds of species. Results over more than a century have documented significant declines in grassland birds, increases in some raptor populations following DDT bans, and range shifts consistent with climate warming. These findings have directly influenced conservation policy and land management decisions across North America.

What These Examples Teach Us

Across all these examples, several patterns emerge. The scientific method is iterative: results often lead to new questions. It is self-correcting: wrong hypotheses are identified and replaced. It values negative results: Semmelweis's data showing what does not cause childbed fever was as important as showing what does. It requires controls and comparisons: without a no-fertilizer group, the plant experiment would be uninterpretable. And it builds on previous work: Hubble built on Slipher's redshift measurements, and Florey built on Fleming's initial observation.