Before you believe something, ask who says it and how they know. A fact is something you can check, while an opinion is what a person thinks, and opinions can be backed up with good reasons even though a reason is not the same as a fact you can check. Whether something is a fact or an opinion does not depend on whether it is right, because a false claim that could be checked is still a claim of fact. Brain-imaging studies, for example, have found no part of the brain that is always switched off. Experts who study the same thing can still have different opinions, so an expert's name alone does not settle a question.

A good source should say where its facts came from. If a claim has no clear source, try to find the same fact somewhere else before you trust it, since a true fact can be found in more than one good place. A page may also say when it was written or last updated, and old facts can go out of date. A first-hand record, like a diary or a letter from the time, is called a primary source, and in science a primary source can be a report in which scientists describe their own research. Secondary sources talk about primary sources: they comment on, sum up or explain what the first-hand records say. Encyclopedias are usually tertiary sources, a collection of primary and secondary sources.

Even a fossil can be a fake. A skull and jaw found in England in 1912 were said to be an early human relative, but the find was a fake: a human skull put with an orangutan jaw. The fossils were found in 1911 and 1912 and included part of a skull, a jawbone and a few teeth, and the jawbone had been treated with chemicals to look as old as the skull, while its teeth were ground down. In 1912 there were no good dating methods, so the claim could not be checked in the way we can now. The skull seemed to match what scientists hoped to find, so many believed it, and the fake fooled scientists for about 40 years before it was proved fake.

Some of the people who looked at Piltdown Man were respected scientists, and they were fooled too. A skull-and-jaw fake was found out in 1953, when scientists tested the bones, and tests showed that the jaw was an ape's. In 1949 the Natural History Museum had a scientist who was not part of the Piltdown discovery test the jaw and skull with fluorine. The hoax slowed the search for human ancestors, because other fossils did not fit it. Tests can show that a claim is wrong.

In 1917 two cousins in a village in England took photos that seemed to show tiny fairies. Many people looked at those fairy photos and believed they were real, even experts at a camera company, and the photos fooled many people for years before the two women who made them admitted in 1983 that they were a hoax. Vikings did not wear helmets with horns when they fought, and we have no proof that they did, because the made-up picture of a Viking in a horned helmet comes from a costume for an opera in 1876, when the first production of Wagner's Ring Cycle used horned helmets in its costumes. That picture has been disproven, but it is still found in modern images.

In 1835 a New York newspaper printed that an astronomer had seen life on the Moon, and the story was made up. It said that a giant new telescope showed the creatures, and it was printed in a cheap newspaper and reprinted by papers around the country. A newspaper that makes a big claim can have a reason to make it, such as selling more copies. Newspaper stories can exaggerate: after a 1938 radio play about Martians, newspapers wrote that many people panicked, but later writers found the panic was much smaller. In 1938 about 5,000 homes were asked what radio show they were listening to, and only 2 percent said the Martian play or one like it.

Some mistakes come from the reader, not the source. Confirmation bias covers how people search for, interpret and remember information so that it fits what they already believe, and it can create self-fulfilling prophecies. In the Piltdown case, scientists wanted to find the first human ancestor in England, and that wish may have made them less careful. Another trap is cause. Two things can rise and fall together without one causing the other: ice cream sales and drownings both rise when the weather is hot. The weather is called a confounding variable here: a third thing that causes both.

Some tests take a real risk. In 1919 astronomers used a total solar eclipse to test whether the Sun's gravity bends starlight, as Einstein's theory said it would. Einstein's theory predicted starlight near the Sun would shift by 1.75 arcseconds, and Eddington's teams measured 1.98 and 1.61. Newton's theory asserted a smaller bend of about 0.87 arcseconds, so the 1919 numbers could tell the two ideas apart. Karl Popper said the impressive thing about the test was the risk: the theory forbade an outcome that could have happened. Falsifiability means a claim could be shown wrong by some possible observation.

Science can change an answer when there is new evidence. Long ago Pluto was called the ninth planet. In 2006 the IAU voted for a new definition of dwarf planet, and Pluto was moved out of the planet group. Dwarf planets were defined as objects that orbit the Sun, are nearly round, but have not cleared their orbit of debris. As of November 2026, NASA lists five dwarf planets, including Pluto.

Words to know

• evidence: checkable proof
• primary: first-hand
• tertiary: collected from others
• falsifiability: can be disproved
• bias: leaning one way