In episode four of the film, Nicholas Ponta advises a student Rebecca to polish her data, partly by ignoring those data and statistical tests that produce unwelcome results. He, moreover, suggests that such behavior is required in order to make an academic career. This is the scientific method. Everybody does it. Now clearly, this is a highly unconventional view. So, let us look at the received view of the so-called scientific method, the method which is held in such high regard. For in fact, it is this method rather than the knowledge it produces, that is generally seen to form the true essence of science. And it is because of this method that scientific knowledge is considered to be far more trustworthy than other kinds of knowledge. So, what is it? What is the secret of science? The scientific method is usually presented as a cyclic or iterative process. It often starts with a question and some preliminary research to find out what is already known about a specific topic. Then, we construct a hypothesis and think of ways to test the hypothesis, usually by doing a critical experiment. The next step is to run the experiment. And finally, we analyze the results of the experiment. If they confirm our hypothesis, we accept the hypothesis to be true, at least, for the time being. If however they reject our hypothesis, then we conclude that the hypothesis is false. In that case, we have to think of a new hypothesis that we can test. And thus, false truths are eliminated and in the end, truth prevails. This at least is the blueprint, the textbook method. Now, many scientists have pointed out that in general this is not how real science works. Some even call the scientific method a myth. In general, real science is much messier than the clear cut picture which we just presented suggests. Scientific research is often chaotic and unpredictable, and projects rarely follow a step by step linear recipe. They often involve repeating the same steps many times to account for new information and ideas. And sight is usually engaged in many different activities, in many different sequences. Actually, some projects not even remotely resemble the method as presented. They may be highly theoretical. That is, they may not involve any experimental tests at all. An example is work on superstrings and theoretical physics, a highly respectable field of science. And some projects are predominantly exploratory. They simply do not involve any hypothesis. Examples are combing a forest looking for new species of plants or animals or sweeping the skies with a new generation telescope. And some projects aim at practical rather than theoretical results. For instance, building a laser or new kind of microscope. Here, the upshot is to create rather than to test something. And then above all, distinguishing truth from falsehood is not as simple as the idealized scientific method suggests. As we already pointed out when discussing Karl Popper, testing a hypothesis is less straightforward than it may seem. Usually, we need to make several additional presuppositions to test an idea, and each of these may be false. And even if they are not, the data may still be flawed as experiments often fail. In fact, one of the most read and cited scientific papers of the last two decades published by John Loannidis, points out that the scientific method, whatever it may be, is highly fallible and far from water proof. In fact the title says it all, Why Most Published Research Findings Are False. As the paper suggests, most published results are chance findings rather than solid facts. In a follow up lecture, I will address some of the causes of the somewhat shocking reality. To end this lecture, let me briefly return to the scientific method. Perhaps, we should regard the mythical scientific method as a moral tale and an injunction. Make sure you know a lot about a topic you want to contribute to. Think deeply about interesting questions. Think hard about a way to translate a question into a testable hypothesis. And finally, in the spirit of sir Karl Popper, use critical tests to learn as much as you can.
