How to Read About Science
A few years ago, I interviewed Daniel Rothman, a professor of geophysics at MIT. We were discussing how hard it is to talk to the public about complicated scientific ideas like climate change. He said, “The world likes simple stories. And this is a problem that does not really lend itself to a simple story.”
Science is important. It fuels new technologies. It explains how the world works. In areas such as medicine and the environment, science can also have a very real impact on people’s lives. We need to know if a new drug will cure cancer, or how climate change will affect sea levels and coastal cities.
Unfortunately, there are some features of science that are not always easy to understand. Good science is often slow and incremental. Near the end of our interview, Dr. Rothman asked, “How do you write new stories that are exciting about something that is simply evolving?”
This makes science hard to write about. And to read about. New experimental results are more interesting to the public (and journalists), but more scientifically uncertain. For every study that makes its way into the mainstream media, countless others are ignored. This distorts the public’s view of the state of science and deprives them of the context necessary to understand what’s going on - context that I hope I can at least in part provide.
Understand That Science is a Process
What is science? Science is one of those words that everyone knows, but it means different things to different people.
Here’s what the Oxford English Dictionary has to say: “The intellectual and practical activity encompassing the systematic study of the structure and behavior of the physical and natural world through observation and experiment.”
The most important word in that definition is “systematic.” Science is not a body of knowledge – it’s a way of doing things. Tom Zeller, editor of Underdark, the digital magazine of MIT’s Knight Science Journalism program, told me in an interview that, “it’s often the case that people think science knows or doesn’t know something. And science isn’t a static basket of knowledge. It’s a process.”
This process is often summed up as the scientific method. When there is a question about the physical world that needs an answer, a scientist begins by hypothesizing what the answer to that question might be. An experiment is designed to test the hypothesis, and the scientist predicts what the results will be. Then the experiment is performed and the results analyzed to see if they match the prediction. And the process is recorded so that at a later date another scientist can attempt to reproduce the experiment’s results.
Karl Popper, a philosopher of science, wrote in The Logic of Scientific Discovery, “it is not his possession of knowledge, of irrefutable truth, that makes the man of science, but his persistent and recklessly critical quest for truth.”
Some people, like Popper, hold that the scientific method needs to be strictly defined. At the other end of the spectrum, there are people like Paul Feyerabend, another philosopher of science, who believe, as he wrote in Against Method, that, “anything goes.”
Many scientists understand, however, that the truth lies somewhere in between. Richard Feynman, the Nobel Prize-winning physicist, once told a group of physics teachers that, “Science is the belief in the ignorance of experts.”
Most science progresses through well-designed procedures. But the history of science shows that science does not always go according to plan. Some scientific discoveries are accidents, like in 1928, when Alexander Fleming, a Scottish bacteriologist, found that a Penicillium mold had infested his bacteria cultures and killed them. He thus discovered penicillin and saved millions of lives. So it’s best to think of the scientific method as more of a guideline than a fixed law.
Become Comfortable With Uncertainty
One implication of the scientific method is that no scientific theory can ever really be proven. There is always an alternative hypothesis than can explain an experiment’s results or observations. If more than one hypothesis is proposed, the simplest theory is usually accepted until they both can be tested. This is not a reason to dismiss a scientific theory out of hand. Even theories that are not completely correct can have value.
For instance, Newton’s equations for mechanics are still useful in almost all practical applications, even though Einstein eventually came along and showed they were not accurate in extreme situations. And there are good reasons to believe that Einstein’s ideas are not entirely accurate either. Scientists are still not sure how quantum mechanics and Einstein’s theory of gravity can work together.
This shows that people have to be open-minded when it comes to science. Everything we know needs to be under constant scrutiny. Joanna Klein, a science reporter for the New York Times, told me in a phone interview that, “You have to be open to the idea it might change.”
However, when I say that science is uncertain, it's important to understand that the level of uncertainty can change drastically depending on the topic. I can say that the earth spins on its axis as it revolves around the sun, therefore the sun will come up tomorrow. From a scientific standpoint, there's an amount of uncertainty in that statement. But from a practical point of view, no one really doubts that the sun is going to rise tomorrow or the day after that.
Understand Sample Size
Daniel Kahneman, the psychologist and Nobel Prize winner in economics, writes in Thinking Fast and Slow that, “The strong bias toward believing that small samples closely resemble the population from which they are drawn is also part of a larger story: we are prone to exaggerate the consistency and coherence of what we see.”
In other words, human beings are bad at statistics. Even scientists don’t know statistics as well as they should. So even if you’re not a math person, there’s one concept you can still keep in mind that will help when you read about science.
Sample Size.
The more measurements or observations a scientist makes, the more accurate an experiment will be. Think about a batter in baseball. Watching one at bat will not give you much of an idea about how good a batter is. You generally need to watch one batter for hundreds of at bats figure it out.
However, what is not as obvious is that the more often an experiment is performed, there is a greater chance that a single experiment will produce a strange result. Every once in awhile a hitter may catch fire and get ten hits in a row. This is amazing since even good hitters will only get three or four hits out of ten chances. But no one thinks that batter is never going to make another out. We recognize it as luck.
However, in science journalism, it is often the case that the only thing you read about is the scientific equivalent of that first at bat or of the batter going ten for ten. These results are new and exciting, but unless you also get the boring reports of the batter going two for ten or three for ten, you get a distorted view of his abilities.
So when the results of scientific studies disagree with each other, it might be a sign that one of them is bad science. But sometimes it’s just a case of the batter going ten for ten versus a case of the batter going zero for ten. The same guy is capable of both outcomes. It just means a lot more experiments might need to be done before we know his true batting average.
For instance, on January 12, Time Magazine published an article, “Tumeric May Not Be a Miracle Spice After All.” Amanda MacMillan writes, “A new review of scientific literature on curcumin, the most well-known chemical in turmeric, suggests that the compound has limited, if any, actual health benefits.” In order to discover this, scientists reviewed over a thousand studies on tumeric. In the past, the media had latched on to the few studies that showed benefits (some of them with questionable methods) while hundreds of other studies that showed no benefit had been ignored.
Putting the Puzzle Together
At this point, you may be asking yourself why, if there is so much uncertainty, we should believe in science at all. I think the answer is, that despite the skepticism necessary to do scientific research, eventually it just works. We’ve developed nuclear power and put men on the moon. But it takes effort and time to develop a theory to the point it can provide tangible results. It’s a tedious process that is usually skipped over by the media in their effort to provide their audience with interesting stories.
Science is like a jigsaw puzzle. One that we have no idea what the finished picture is going to look like. Each experiment or observation is another piece of the puzzle. When we only have a few pieces, it’s difficult to guess what the picture will be. This is why it is dangerous to put too much stock into stories about new science. It’s like guessing the picture of a puzzle by looking at a single piece.
Once you have enough of the pieces, though, you can guess (hypothesize) what the whole picture will be. Based on that guess you can predict what another piece will look like. When you find that piece (through experiment), it will either match the picture you imagined or it won’t. Then you revise your idea of the picture.
This shows why, despite all the uncertainty, science eventually works. At a certain point, we have enough pieces that the picture begins to emerge clearly and we can all agree on what we see.