Description of the Scientific Method

I have been making my way through Zen and the Art of Motorcycle Maintenance and have just encountered a good description of the scientific method. Chapter 9 describes two types of logic (inductive and deductive) that are applied in a broader framework in an attempt to unambiguously solve a problem.

Inductive logic is described as starting from observations and inferring a general conclusion. Deductive logic is described as the reverse as we start with general knowledge of a system and predict a specific observation.

The use of these reasoning tools need to be applied in a formal context so that any results obtained can be defended. This is the scientific method and the book breaks the method down to six steps:

1. Statement of the problem;
2. Hypotheses as to the cause of the problem;
3. Experiments designed to test each hypothesis;
4. Predicted results of the experiments;
5. Actual results of the experiments;
6. Conclusions from the results of the experiments.

One key point when working through this process is to not jump the gun when stating the problem. The problem statement must not imply a cause of the problem. The statement should be as general as possible and then possible causes can be stated in step 2 and tested in the subsequent steps. Once a particular hypothesis is proved to be true, it can then form the problem statement in a new round of experimentation to further isolate the source of the problem.

This art of correctly stating a problem is also raised in Conceptual Blockbusting: A Guide to Better Ideas where it is stated that an incorrect problem statement can impact on the creative process by limiting the number and type of solutions generated.

I wish I had a better understanding of these ideas seven years ago when I started my PhD!