Scientific Investigation and Experimental Design

The 100-Year Fever Mystery

In 1847, a doctor named Ignaz Semmelweis noticed something horrifying. Women in one maternity ward of his Vienna hospital were dying of fever at five times the rate of women in the ward next door. Same hospital. Same city. Same year. What was different?

Semmelweis tracked the data obsessively. He tested hypothesis after hypothesis. Was it the air? The food? The position women delivered in? Nothing explained it.

Then a colleague died after cutting his finger during an autopsy, and Semmelweis noticed the symptoms matched the fever killing the mothers. The doctors in Ward 1 performed autopsies in the morning, then delivered babies in the afternoon — without washing their hands. Ward 2 was staffed by midwives who never touched cadavers.

Semmelweis ordered doctors to wash their hands in a chlorine solution before examining patients. The death rate plummeted from 18% to under 2%.

The medical establishment rejected his findings. They were insulted. A gentleman's hands couldn't possibly carry disease — that was absurd. Semmelweis died in an asylum in 1865, decades before germ theory proved him right.

His story is a perfect case study in how science works: observation, hypothesis, experiment, data, conclusion — and how human stubbornness can resist even the clearest evidence.

Hypotheses Are Not Guesses

A hypothesis is not a guess. A guess is "I think plants like music." A hypothesis is a specific, testable prediction grounded in reasoning: "If bean plants are exposed to 30 minutes of classical music daily, then they will grow 15% taller than silent controls over 21 days, because sound vibrations may stimulate cell growth."

Notice the structure: If (what you'll do) then (what you predict will happen) because (your reasoning). The "because" is what elevates a hypothesis from a dart thrown at a board to an informed prediction rooted in existing knowledge.

A good hypothesis must be falsifiable — meaning an experiment could prove it wrong. "Plants have feelings" is not falsifiable because no experiment could definitively disprove it. "Plants exposed to music grow taller" is falsifiable because you can measure it and get a clear yes or no.

❝

"In so far as a scientific statement speaks about reality, it must be falsifiable; and in so far as it is not falsifiable, it does not speak about reality."

Variables: The Engine of Good Experiments

Every experiment revolves around three types of variables:

• Independent variable: What you deliberately change. You control this. (Amount of fertilizer.)
• Dependent variable: What you measure to see if it changed. This responds to what you did. (Plant height.)
• Controlled variables: Everything you keep identical so it doesn't contaminate your results. (Same soil, same water amount, same light, same pot size, same plant species.)

Here is the brutal truth about bad experiments: if you change two things at once, you learn nothing. If you give one plant more fertilizer AND more sunlight, and it grows taller, which caused it? You have no idea. You've wasted your time.

The Texas State Science Fair sees this mistake constantly. A student "tests whether organic food is healthier" but compares organic apples to conventional bananas. Different fruit, different growing conditions, different everything. No conclusion possible.

Measurement: Precision Is Not Optional

Science speaks in numbers, and those numbers must be honest. In the United States, we use inches and pounds in daily life, but science worldwide uses the SI system (International System of Units) so that a measurement in Houston means the same thing in Hamburg.

Significant figures matter. If your ruler has millimeter markings, you can estimate to 0.1 mm — but claiming you measured 3.1472 mm is dishonest precision. Report what your instrument can actually resolve, nothing more.

Pseudoscience: The Lab Coat Costume

Every day you encounter claims that sound scientific but crumble under scrutiny. "Magnetic bracelets cure arthritis." "Mercury retrograde affects your decisions." "This detox juice flushes toxins."

How do you tell real science from pseudoscience wearing a lab coat? Ask five questions:

1. Is the claim testable? Can you design an experiment to check it?
2. Has it been replicated? One study proves nothing. Independent replication is everything.
3. Who benefits from you believing this? Follow the money.
4. Does it rely on anecdotes or data? "My aunt tried it" is not evidence.
5. Does it update when challenged? Real science changes with new evidence. Pseudoscience doubles down.

Why Science Changes Its Mind

People sometimes distrust science because it "keeps changing." But updating conclusions when new evidence appears is the entire point. That is not weakness — it is the mechanism that makes science self-correcting.

Doctors once prescribed bloodletting for infections. Geologists once believed continents were fixed in place. Physicists once believed light needed a medium called "ether" to travel through. Every one of these ideas was replaced when better evidence emerged.

The fact that science can admit it was wrong and correct course is precisely what makes it trustworthy. Any system that claims to never make mistakes is lying to you.

❓

Concept Check

Explain the difference between a hypothesis and a theory in scientific terminology. Why is 'just a theory' a misunderstanding of how science uses the word?

▸

A hypothesis is a testable prediction about a specific outcome — it hasn't been extensively tested yet. A scientific theory is a broad explanation that has been tested repeatedly, supported by massive evidence from multiple independent studies, and has never been disproven. 'Just a theory' misunderstands the word because in everyday language, 'theory' means a guess. In science, a theory (like the theory of gravity or cell theory) represents our most robust, well-supported understanding of how something works. Theories are the highest level of scientific confidence, not the lowest.

❓

Concept Check

Semmelweis's hand-washing data showed death rates dropped from 18% to under 2%, yet doctors rejected his findings for decades. What does this reveal about how science actually advances versus how we imagine it advances?

▸

It reveals that science does not advance purely through logic and evidence — human psychology, ego, and institutional resistance play enormous roles. We imagine scientists as perfectly rational beings who accept evidence immediately, but real scientists have biases, reputations to protect, and emotional investments in existing ideas. Semmelweis's case shows that strong evidence is necessary but not sufficient — it also takes time, persistence, and often a generational shift for paradigm-changing ideas to be accepted. Science is self-correcting, but the correction can take decades.