Essay on the Various Scientific Method



Everything that is science is ultimately based on some scientific method. Taken singly, most of the steps of such a method involve commonplace procedures carried out daily by every person.

Taken together, they amount to the most powerful tool man has devised for learning about nature and making natural processes serve human purposes.

Observation, Problem and Hypothesis

Science generally begins with observation, the usual first step of scientific inquiry. This step immediately limits the scientific domain; something that cannot be observed, directly, or indirectly, cannot be investigated by science.

Furthermore, for reason that will become clear presently, it is necessary that an observation be repeatable, actually or potentially. One-time events are outside science (the one-time origin of the universe possibly accepted).

Correct observation is a most difficult art, acquired only after long experience and many errors. Everyone observes, with eyes, ears, touch, and all other senses, but few observe correctly. The problem here is largely unsuspected bias.

People forever see what they want to see or what they think they ought to see. It is extremely hard to rid oneself of such unconscious prejudice and to see just what are actually there, no more and less. Past experience, "common knowledge", and often teachers can be subtle obstacles to correct observation, and even experienced scientists may not always avoid them.

That is why a scientific observation is not taken at face value until several scientists have repeated it independently and have reported matching data. That is also a major reason why one-time, unrepeatable events generally cannot be investigated scientifically.

After an observation has been made, a second usual step of scientific procedure is to define a problem; one asks a question about the observation. "How does so-and-so come about?" "What is it that makes such-and-such happen in this or that fashion?"

Question-asking again distinguishes the scientist from the layman; everyone makes observations, but not everyone has the curiosity to go further. Indeed, few become aware that a particular observation actually might pose a problem.

For thousands of years, even curious people simply took it for granted that a detached, unsupported object falls to the ground. It took genius to ask "How come?" and not many problems have ever turned out to be more profound.

Thus, scientists take nothing for granted and they ask questions, even at the risk of irritating others. Question- askers are notorious for getting themselves into trouble, and so it has always been with scientists. But they have to continue to ask questions if they are to remain scientists, and society has to expect annoying questions if it wishes to have science.

Like good observing, good questioning is a high art. To be valuable scientifically, a question must be relevant as well as testable. Often it is difficult or impossible to tell in advance whether a question is relevant or irrelevant, testable or untreatable. If a man collapses on the streets and others want to help him, it may or may not be irrelevant to ask when the man had his last meal.

Without experience one cannot decide on the relevance of this question, and a wrong procedure might be followed. As to testability, it is clear that proper testing techniques must be available. But this cannot always be guaranteed.

For example, Einstein achieved fame for showing that it is impossible to tell whether or not the earth moves through"ether", an assumption held for many decades. All questions about ether therefore became untreatable. Einstein reformulated such questions and came up with relativity, an idea that posed fully testable problems.

In general, science does best with "how" or "what" questions. "Why" questions are more troublesome. Some of them can be rephrased to ask "How?" or "What?" But others, such as untreatable category. These are outside the domain of science.

Once a proper question has been asked, the third step of scientific methodology usually involves the seemingly quite unscientific procedure of guessing what the answer to the question might conceivably be. Scientists refer to this as postulating a hypothesis.

Hypothesizing distinguishes the scientist still further from the layman. For while many people observe and ask questions, most stop there. Some do wonder about likely answers, and scientists are among these.

Since a particular question usually has thousands of possible answers but in most cases only a single right one, chances are excellent that a random guess will be wrong. The scientist will not know if his guess was or was not correct until he has completed the fourth step of scientific inquiry, experimentation.

It is the function of experiments to test the validity of scientific guesses. If experiments show that a first guess was wrong the scientist then must formulate a new or modified hypothesis and perform new experiments. Clearly, guessing and guess-testing could go on for years and a right answer might not be found. This happens.

But here artistry, genius, and knowledge of the field usually provide shortcuts. There are good guesses and bad ones, and the experienced scientist is generally able to decide at the outset that, of a multitude of possible answers, so-and-so many are unlikely answers.

This is also the place where hunches, intuitions, and lucky accidents aid science enormously. The ideal situation the scientist will strive for is to reduce his problem to just two distinct alternative possibilities.

Experimental tests should then answer one of these with a clear "yes", the other with a clear "no". It is exceedingly difficult to streamline problems in this way, and with many it cannot be done. Very often the answer, obtained is "maybe". But if a clear yes or no does emerge, the result well might be a milestone in science.

Experiment and Theory

With the next general step in a scientific inquiry, experimentation, science "and conscience part company completely. Most people observe, ask questions, and also guess at answers. But the layman then stops; "My answer is so logical, so reasonable, and it sounds so 'right', that it must be correct."

The listener considers the argument finds that it is indeed logical and reasonable, and is convinced. He goes out and in his turn converts others. Before long, the whole world rejoices that it has the answer.

Now the small, kill-joy voice of the scientist is heard in the background. "Where is the evidence?" Under such conditions in history it has often been easier and more convenient to ignore the scientist that to change emotionally fixed the scientist than to change emotionally fixed public opinions. But disregarding the scientist does process must be repeated until a hypothesis is hit upon that can be supported with confirmatory experimental evidence.

As with legal evidence, scientific evidence can be strong and convincing, or merely suggestive or poor. In any case nothing has been "proved". Depending on the strength of the evidence, one merely obtains a basis for regarding the original hypothesis with a certain degree of confidence.

Our new drug, for example, may be just what we claim it to be when we use it in this country. In another part of the world it might not work at all or it might work better. All we can confidently say is that our evidence is based on so-and-so many local experiments, and that we have shown the drug to have an effectiveness of 50 per cent. Experimental results are never better or broader than the experiments themselves.

This is where many who have been properly scientific up to this point become unscientific. Their claims exceed the evidence; they mistake their partial answer for the whole answer; they contend that they have "proof for a "fact", while at they actually have in some evidence for a hypothesis. There is always room for more and better evidence, or for new contradictory evidence, or indeed for better hypotheses.

Experimental evidence is the basis for a fifth general step in scientific procedure, the formulation of a theory. In our drug example, a simple theory would be the statement that "against such-and-such a bacterial disease, drug X is effective in 50 per cent of the cases".

To be sure, this statement cannot be regarded as a particularly significant or far-reaching theory. Nevertheless, it implies, for example, that drug X will be 50 per cent effective anywhere in the world, under any conditions, and can be used also for animals other than man. Direct evidence for these extended implications does not exist.

But inasmuch as drug X is already known to work within certain limits, the theory expresses the belief, or probability, that it will also work within certain wider limits. To that extent every good theory has predictive value; it forecasts certain results.

In contrasts to nonscientific predictions, scientific ones always have a substantial body of evidence to back them up. Moreover, a scientific forecast does not say that something will certainly happen, but says only that something is likely to happen with a stated degree of probability.

A few theories have proved to be so universally valid and to have such a high degree of probability that they are spoken of as natural laws. For example, no exception has ever been found to the observation that an apple disconnected from a tree and not otherwise supported will fall to the ground.

A law of gravitation is based on such observations. Yet even laws do not pronounce certainties. For all practical purposes it well might be irrational to assume that someday an apple will rise from a tree, but there simply is no evidence that can absolutely guarantee the future. Evidence can be used only to estimate probabilities.

Most theories have rather brief life spans. For example, if our drug X should be found to perform not with 50 percent but with 80 percent efficiency in chickens, then our original theory becomes untenable and obsolete. The exception to the theory now becomes a new observation, the start of a new cycle of scientific investigation.

New research might show, for example, that chickens contain a substance in their blood that enhances the action of the drug substantially. This finding might lead to isolation, identification, and mass production of the booster substance, hence to worldwide improvement in curing the bacterial disease. And we would also have a new theory of drug action, based on the new evidence.

Thus, science is never finished. One theory predicts holds up well for a time, exceptions are found, and a new, more inclusive theory takes over-for a while. Science is steady progression, not sudden revolution. Clearly, knowledge of scientific methodology does not by itself make a good scientist, any more than knowledge of English grammar alone makes a Shakespeare.

At the same time, the demands of scientific inquiry should make it evident that scientists cannot be the cold, inhuman, precision, machines they are so often and so erroneously pictured to be. Scientists are essentially artists who require sensitivity of eye and of mind as great as that of any master painter, and an imagination and keen inventiveness as powerful as that of any master poet.