ABSTRACT
When people think of natural scientists doing research, they tend to think of them in a laboratory, probably doing an experiment. The experiment is the classic research method of the natural scientist, and has produced many of its most valuable results, both in pure and in applied science. This is because the experiment is most suited to the assumptions that natural scientists have traditionally made about what they are studying. They assume that the natural world has an independent existence of its own, which is as it is regardless of those who are studying it, and which is governed by laws which can be discovered by the research scientist if only the right methods can be developed. The knowledge that is discovered using these methods is regarded as objective and factual, i.e. it is correct for all times and all places, and is not going to be different according to who discovers it. Once that knowledge is gained, it can be used to explain events in the natural world, to make predictions about what will happen in the natural world, and thus to control that world and make it behave in ways that are, at least in theory, to the advantage of the controllers. We will pursue this debate further in chapter 6 of this book, and find that this picture of natural science has been modified recently. For the moment, however, we will accept these assumptions,
The hypothetico-deductive method
This is the name given to the logic of the research method that natural science is thought to employ. It is illustrated in Figure 3. The process starts with the phenomena (1), out there in the world, which can be observed objectively. These sometimes casual observations prompt ideas in the mind of the scientist (2), from which is developed an hypothesis (3). This is essentially an intelligent guess about what is happening, but in a form that can be tested. This test may be done as an experiment in the laboratory, or by further data collection ‘in the field’. The scientist should try to prove the hypothesis wrong. This is because, however often an experiment comes out right, you can never be sure that it will come out right the next time, and so you can never be sure that your hypothesis is right. It only has to come out wrong once for you to know that the hypothesis is wrong. An experiment is then carried out (or data is collected in the field) (4), and the results are analysed (5). The hypothesis is then tested against these results (6). If the hypothesis is not supported by the evidence (7), it can be rejected or revised (8), and a fresh hypothesis developed (3). If it is supported by the evidence (9), then it can be seen as a contribution to theory (10). If the theory becomes elaborate enough, it begins to look like a law, and predictions about future events may be made (11).
