Desirable Qualities of Hypotheses

The use of hypotheses are as widespread and their varieties are large. Isaac Newton’s Laws of Motion were hypotheses he set himself to prove. Finding my lost set of keys also requires hypotheses. Some qualities serve as criteria to distinguish great hypotheses from trivial ones and, to an extent, to judge the soundness of a given hypothesis. Hypothesizing is part of human thought—we all do it in our daily lives. But to produce a hypoth­esis of the stature of the Laws of Motion is an extraordinary event, even among great scientists.

All those who have a Ph.D. in philosophy are not philoso­phers; all those who have a Ph.D. in literature are not poets or writers. (In fact, great creative writers seldom have Ph.D’s) But most scientific researchers—experimental or theoretical—have had the experience, however slightly, of forming scientific hypotheses. In fact, in the course of research, let us say the three years spent by a Ph.D. candidate, the most crucial time, the turn­ing point, is the event of formulating the hypothesis. Once that is done, in a sense, the rest is routine: doing experiments, writing the theses, receiving the degree, and, of course, looking for a job.

The excitement of incubating and hatching a hypothesis is quite a feeling; it has come, and it is gone! During that period or soon after, it is worthwhile for the researcher to spend some time analyzing the hypothesis itself to make sure that it is not a fake. In these days of group research—with the emphasis on team- work—it is possible that the individual researcher will be given a hypothesis and asked to use his skill and knowledge to do the experiments, then create a report or a thesis. Such an individual is lucky in that he avoids a lot of frustration, akin to searching for an unrecognizable object in a dark room, and dead-end literature studies. But he is also deprived of the worthwhile excitement of forming the hypotheses in his head and carrying it through to confirmation, working like an artist, giving shape to his dream!

Whether a group or an individual forms the hypothesis, sub­jecting the hypothesis to scrutiny is desirable. The following are some well-known qualities of a good hypothesis:

1. The first and foremost purpose of a hypothesis is to answer the specific question at hand as closely and directly as possible, without bringing into the domain of inquiry unnecessary details and imper­tinent connections. For example, if our investiga­tor on electrical resistance set himself the task of shining the wires to absolute cleanliness or finding the metallurgical composition of each wire to the accuracy of the third decimal, he might be perfect­ing some possible variables, but he would be lead­ing himself away from the problem at hand.
2. A hypothesis should be testable, and the results of the test should be observable in terms of material objects and entities, as physical or chemical attributes or measurable qualities or changes, which can be directly perceived. If a psychic person claims that he can see ghosts and tells us their size, but asserts that others cannot see them and that the size cannot be found by ordinary measuring instru­ments, then the existence of ghosts, stated as a hypothesis, is useless because nobody can confirm this assertion by means of experiments.
3. A hypothesis should have deductive capacity. Sup­pose a hypothesis is confirmed by means of a set of tests or experiments; the confirmed hypothesis then becomes a part of the body of science. This fact should not exhaust the capacity of the con­firmed hypothesis (from then on referred to as a theory) to deduce one or more consequences of itself that will also be true. Each such conse­quence, in effect, becomes a new hypothesis to be tested. To the extent that a deduced hypothesis also passes the test, the original hypothesis gains credibility. A well-known example in recent times is the confirmation test done for the Theory of Relativity. A deduction from the theory was that light, like any material object, is subjected to grav­itation. Though the theory was presented in 1905, it was not fully accepted by the scientific commu­nity until the test was done in 1919, and the result was found to be positive. The greater the deduc­tive capacity of a given hypothesis, the better the hypothesis. We should note, however, that deduc­tive capacity may appear similar to, but is quite different from, testability.

4. A new hypothesis should be consistent with the existing, established body of science in that field. Most often the contribution made by new research, say doctoral work, is like adding another brick to the growing wall for a room in a big building under construction. The brick should “form a good fit” to be acceptable and useful as a modest part of the building. But there are excep­tions to this rule; nonconformity is not necessarily always a defect. Very rarely, it may happen that a hypothesis, by not fitting well, will call for an entirely new structuring of science—a whole new building! For example, the principle of the conser­vation of mass, which is still taken for granted in most physics, chemistry, and technology calcula­tions, had to be given up on strict consideration when mass and energy were shown to be equiva­lent in the early part of this century. Now, we have, instead, the principle of the conservation of energy, wherein mass is simply a component. Is the new conservation principle true for all time? So far, yes, but there may be nothing that lasts for all time in science.
5. A simple hypothesis is better than a complex one. As an artist could say, “Simplicity is beauty”; so could a scientist. But the answer to the question, What is simplicity? may not be obvious. The artist may venture to explain: instead of painting a young female model wearing elaborate clothing and jewelry, he prefers to paint her nude because the latter arrangement, to him, is simple. But it may not be so easy for a scientist to find an equiv­alent “simple” arrangement to convince himself that his hypothesis is simpler than a rival hypothe­sis. Simple testability (meaning the requirement of a simple and direct test or experiment for confir­mation), the capacity to beget one or more deduc­tions that can in turn be used as crucial tests, and the ability to explain one or more prevailing puz­zles in science are some attributes of a simple, and superior, hypothesis.

Source: Srinagesh K (2005), The Principles of Experimental Research, Butterworth-Heinemann; 1st edition.