Your goofy but lovable cousin just told you that you should stop eating eggs because he read somewhere that a study showed they are bad for you.
How much should you trust your relative on such matters? More importantly, how much should you rely on one newly published bit of research when deciding what to make for breakfast?
To be clear, this is not an article about the health-promoting or health-torpedoing properties of eggs. It’s about how scientific knowledge is built piece by piece from many studies. What scientists know is refined over time as new results either do or don’t point to the same conclusion.
I’m a geographer who’s been doing and teaching science for many decades, with a sideline of teaching and writing about how science is done. Many people, quite understandably, take a single experiment or study as the be-all and end-all of knowledge because that’s how research often is presented by the press or on social media. But the better way to approach new research is to find how it weaves together with other work on the topic to create big-picture understanding.
Science evolves over time as more data and discoveries refine scientific knowledge.
Historica Graphica Collection/Heritage Images via Getty Images
How science works
Most research studies are undertaken either to fill a gap in our knowledge or to test an existing theory to see whether it deserves the confidence people have in it. After identifying the topic, scientists design a study to achieve those ends. They may run an experiment to learn more about how a chemical affects certain cells, for instance, or collect data in the field to track a natural phenomenon, such as how water temperatures affect hurricanes.
Then the researchers submit their findings to a peer-reviewed journal, where other experts – the scientists’ peers – decide whether it’s quality research deserving of publication.
Not all journals have rigorous peer review. Papers are highly unreliable if published by “paper mills” – journals that appear scholarly but will publish anything if the authors pay a fee.
Peer review doesn’t guarantee that the conclusions are valid, but it increases the chances that they are. Individual papers might be wrong because of honest mistakes, such as unforeseen limitations in the experimental design or, rarely, from outright fraud.
No scientific paper solves a problem once and for all. Neither does it negate all previous research. Well-done research contributes a bit to the scientific community’s understanding of a topic. The next, and crucial, step is putting individual studies in context with other research on the topic.
Even if there is current consensus, a new study may reveal a weakness, and that could lead to more research to figure out what is more likely to be correct. Scientific knowledge is constantly being refined as new information comes to light.
Adding more evidence bit by bit
One…
