Astrology and astronomy both study the sky. Only one of them changes its claims when the evidence says it should. Learning to tell the two apart is one of the most useful skills science can teach.
Astrology and astronomy both look at the sky. Homeopathy and pharmacology both talk about treating illness. In each pair, one field follows the scientific method and the other borrows the language of science while avoiding the parts that could prove it wrong. Learning to spot that difference protects you from false claims, wasted money and sometimes real harm. This article gives you the tools to do it fairly.
What science is · what pseudoscience is · testability · warning signs · thinking traps · fair tests · peer review · the evidence ladder · case studies · three worked examples · key terms · common mistakes · practice questions and a quiz.
Science is not just a list of facts. It is a way of finding out about the natural world: asking questions, making testable explanations, collecting evidence, and changing our minds when the evidence demands it. If you have read our article on the Scientific Revolution, you have seen how this habit slowly became normal.
Pseudoscience (the word means "false science") refers to claims, beliefs or practices that are presented as scientific but do not follow the scientific method. They are not genuinely testable, they are not open to being proven wrong, and they are not required to change when evidence contradicts them.
It helps to separate three different things that are often mixed up:
The line between them is not always sharp. Philosophers of science still debate exactly how to draw it, and no single test settles every case. What we can do is look at several signs together.
A scientific claim must be testable: there must be some possible observation that could show it is wrong. The philosopher Karl Popper (20th century) made this famous under the name falsifiability. "Falsifiable" does not mean "false". It means that the claim risks being shown false by evidence. A claim that can never be wrong, whatever happens, tells us nothing about the world.
Claim A: "On 12 March, the Moon will be in exactly this part of the sky at 9 pm."
Claim B: "Good fortune is coming your way, though it may take a different shape than you expect."
Step 1. Ask: what result would show the claim is wrong?
Step 2. For A, we can look up and see whether the Moon is there. If it is not, A is wrong. A is testable.
Step 3. For B, almost anything that happens can be called "good fortune in a different shape". No result could prove it wrong. B is not testable.
Answer: A is a scientific-style claim; B is the kind of vague statement found in many horoscopes.
This is why astronomy is a science and astrology is not. Astronomers can predict the positions of planets years ahead using physics, and a telescope can check them. Astrology says that planetary positions shape personality and events, but its statements are usually too loose to fail a test. When astrology has been tested in careful ways, it has not held up. In a well-known 1985 experiment published in the journal Nature, practising astrologers tried to match birth charts to personality profiles, and they did no better than chance.
No single warning sign proves a claim is pseudoscientific. But when several appear together, it is a strong signal to look closer before accepting it.
| Warning Sign | What It Looks Like |
|---|---|
| Untestable claims | No experiment could ever show the claim is wrong |
| Relies on anecdotes | "It worked for my cousin" replaces controlled evidence |
| Ignores contrary evidence | Failed results are dismissed rather than investigated |
| Cherry-picking | Only the results that fit are shown; the rest are hidden |
| Explaining away failure | "It only works if you believe in it" or "the test disturbed the energy" |
| No peer review | Claims go straight to the public, skipping expert scrutiny |
| Secret or ancient knowledge | "They don't want you to know this" replaces open evidence |
| Cure-all promises | One product or method is said to fix a huge range of problems |
| Jargon without meaning | Scientific-sounding words ("energy", "quantum", "toxins") used without clear definitions |
Falling for a false claim does not mean someone is foolish. Human minds have habits that make weak evidence feel strong. Knowing them helps you catch yourself.
Observation: Across many towns, months with higher ice cream sales also have more drowning accidents.
Step 1. Ask whether ice cream could cause drowning. There is no sensible mechanism for that.
Step 2. Look for a third factor that affects both. Hot weather makes people buy more ice cream, and it also makes more people swim.
Step 3. Conclusion: the link is explained by the hidden factor (temperature), not by ice cream.
Lesson: Before accepting "A causes B", look for a third factor, and ask whether a fair test has been done.
To test whether a treatment, fertiliser or teaching method really works, scientists use a controlled experiment. The key features are:
Claim: "Super Grow spray makes bean plants grow taller."
Step 1: Make groups. Plant 20 identical bean seeds in the same soil. Randomly choose 10 for the spray (treatment) and 10 for plain water (control).
Step 2: Keep everything else equal. Same pot size, same light, same watering schedule. Only the spray differs.
Step 3: Reduce bias. Label the pots with codes so whoever measures the plants does not know which are which.
Step 4: Measure. After 3 weeks, measure the height of every plant and compare the average heights of the two groups.
Step 5: Repeat. If the result is real, other people repeating the test should find a similar difference.
Before most scientific findings are accepted, they pass through peer review, a process where other experts in the same field check the study's methods and evidence before it is published.
Peer review is not perfect. Reviewers can miss mistakes, and a few published studies have later been withdrawn (this is called a retraction). Passing peer review does not make a claim true; it means the work has had expert scrutiny. What makes science trustworthy is that the whole system keeps checking: errors are found, reported and corrected in the open.
Not all evidence is equally strong. As a rough guide, the more fairly a claim has been tested, and the more often independent teams have confirmed it, the more we can trust it.
Phrenology. In the early 1800s, Franz Joseph Gall and his followers claimed that the bumps on a person's skull revealed their talents and character. It was popular for decades. As brain science developed, researchers found that the skull's shape does not map onto traits in this way, and the idea was abandoned. It shows how a confident, scientific-sounding idea can still be wrong.
Lysenkoism. In the Soviet Union from the 1930s, the agricultural claims of Trofim Lysenko were backed by the government while mainstream genetics was rejected. Many scientists who disagreed lost their jobs or were punished. The episode is widely cited as an example of ideology overriding evidence, with damaging results for farming and for science.
A useful counter-example: continental drift. In 1912 Alfred Wegener proposed that the continents slowly move. Many scientists rejected it at first because he could not explain a mechanism. Decades later, new evidence from the ocean floor led to the theory of plate tectonics, and the basic idea was accepted. This shows that being unusual is not the same as being pseudoscience. Strange ideas become accepted when the evidence arrives, and pseudoscience is marked by avoiding that test.
Many homeopathic remedies are diluted so many times that, by calculation, not even one molecule of the original substance is likely to remain in the final product. When the evidence from controlled trials has been reviewed, major health reviews have concluded that there is no good-quality evidence that homeopathy works better than a placebo for any health condition. Even so, some products are still sold with scientific-sounding language. This follows the pattern in this article: familiar vocabulary, but an unwillingness to accept what repeated testing shows. This is general education, not medical advice; if you are unwell, speak to a qualified doctor.
When you meet a claim in an advert, a video or a message, run it through these five questions.
Advert: "The Brain-Boost Bracelet balances your body's energy and raises exam marks. Thousands of happy customers!"
Step 1: Testable? "Balances your energy" is not defined, so it cannot be measured. Red flag.
Step 2: Specific predictions? "Raises exam marks" could be tested, but no numbers are given. Partly vague.
Step 3: Fair tests? The only evidence is "happy customers", which are anecdotes without a control group. Red flag.
Step 4: Independent confirmation? No studies are named. Red flag.
Step 5: Open to change? No mention of what result would make the sellers admit it does not work. Red flag.
Conclusion: Several red flags together, so treat the claim as unproven until a fair test supports it.
| Science | Pseudoscience | |
|---|---|---|
| Testability | Makes specific, checkable predictions | Claims are vague or impossible to check |
| Response to failed tests | Hypothesis is revised or rejected | Failed results are explained away or ignored |
| Evidence used | Controlled experiments and measurements | Stories, testimonials, selected examples |
| Review process | Checked by independent experts before publication | Often announced directly to the public |
| Attitude to new evidence | Open to being proven wrong | Resistant to changing its claims |
| Term | Meaning |
|---|---|
| Hypothesis | A testable explanation that has not yet been confirmed |
| Falsifiable | Able, in principle, to be shown wrong by evidence |
| Control group | The comparison group that does not receive the thing being tested |
| Placebo | An inactive treatment that looks like the real one |
| Double-blind | Neither participants nor researchers know who is in which group until the end |
| Peer review | Expert checking of a study before publication |
| Replication | Repeating a study independently to see whether the result holds |
| Anecdote | A personal story offered as evidence |
| Confirmation bias | Noticing only the evidence that agrees with what you already think |
Try each question first, then tap it to reveal the answer.
A claim is falsifiable if there is some possible observation or result that could show it is wrong. It does not mean the claim is false.
It gives a fair comparison. Without it, we cannot tell whether improvement came from the treatment or from other things such as the placebo effect or natural change.
Neither the participants nor the researchers measuring the results know who is in the treatment group and who is in the control group until the end.
No. Testimonials are anecdotes with no control group, and unhappy customers may not be shown.
No. A third factor, sunny weather, affects both. This is correlation without causation.
Some unusual ideas, such as continental drift, were later accepted when evidence arrived. Pseudoscience is marked by avoiding real tests, not by being new or strange.
This topic works best as an activity. Try the Barnum test: write one general paragraph (for example, "You are usually friendly but sometimes need time alone, and you want others to respect your effort"), give every student the same text as "a personal profile", and ask them to rate how well it fits before revealing it was identical for everyone. A second activity is the bean-plant fair test in Worked Example 3. Discussion prompts: What evidence would change your mind about a belief you hold? Who benefits when a product claim goes unchecked? This article is general education for students up to age 18. It is not medical advice, and it aims to teach how to weigh evidence rather than to mock any group of people.
10 questions. Select an answer for each, then submit to see your score instantly.