Testability

This article is about hypothesis testing. For the ability of equipment to be tested, see Non-functional requirement and Software testability.

Testability, a property applying to an empirical hypothesis, involves two components:

1. Falsifiability or defeasibility, which means that counterexamples to the hypothesis are logically possible.
2. The practical feasibility of observing a reproducible series of such counterexamples if they do exist.

In short, a hypothesis is testable if there is a possibility of deciding whether it is true or false based on experimentation. This allows to decide whether a theory can be supported or refuted by data. However, the interpretation of experimental data may be also inconclusive or uncertain.

Verification and falsification

In philosophy, Verificationism is one of the elements of testability. The principle of verification, established by Sir A. J. Ayer, states that a given proposition, if it cannot be verified by the senses, is meaningless. Falsifiability, a principle of testability introduced by Sir Karl Popper, puts forth that a hypothesis that could be easily falsified, but has not been, is preferable to other hypotheses. The commonality between these two principles is the assertion that a proposition is only good insofar as it is testable. In Popper's method, a given hypothesis will be verified by consistent failure to prove it false, while in Ayer's method, a proposition will be verified by the ability to measure and experience it through one's own senses.

Popper's falsifiability principle, as he intended, works well as a standard for accepting scientific theories. Though it does not fully conquer the problem of induction, Popper's tool establishes the criteria for deciding which hypothesis, of a given selection of hypotheses, is most likely to be accurate. Falsifiability's main tenet is the idea that a hypothesis can be disproven by a single counterexample, and a hypothesis that could very easily be refuted, but has endured nonetheless due to no known instances of such a counterexample, is the superior hypothesis. In this sense, it is similar to Occam's razor, which states that simplicity is to be preferred to complexity, and that entities should not be multiplied beyond necessity.

Ayer's principle of verification, because it posits that any unverifiable claim is worthless, could be taken as cause to reject such schools of thought as metaphysics and ethics, which are not tangible to the senses. In fact, by its own rules, the principle of verification cannot be verified, which results in reductio ad absurdum and renders the principle unreliable. However, it can still be useful when evaluating propositions as it can be used as a basis for rejecting unfounded claims.

See also

• Confirmability
• Controllability
• Observability
• Scientific method
• Test method

Further reading

• Johansson, Lars-Goran (2015). "Hypotheses and hypothesis testing". Philosophy of science for scientists. Cham: Springer-Verlag. pp. 41–61 (59). doi:10.1007/978-3-319-26551-3. ISBN 9783319265490. OCLC 923649072. ... the question of whether the auxiliary assumption is testable or not is not so easy to determine as it might first appear. Criteria regarding independent testability do not seem to be absolute. ... The least common denominator for all sciences is that hypotheses are formulated and tested. This is meaningful only if one is prepared to change one's mind after testing, to admit that even one's favourite hypothesis was wrong. ... The result of the test is either that the predictions and observation reports are compatible, or that they conflict. In the former case one may be justified to say that one's hypothesis is supported. In the latter case one must reconsider something; one must reject either the hypothesis, some auxiliary assumption, or the observation report.
• Kegan, Robert; Lahey, Lisa Laskow (2009). "Designing tests of your big assumption". Immunity to change: how to overcome it and unlock potential in yourself and your organization. Leadership for the common good. Boston: Harvard Business Press. pp. 256–264. ISBN 9781422117361. OCLC 231580325. The purpose of each test you run is to see what happens when you intentionally alter your usual conduct and then reflect upon the meaning of the results for your big assumption. ... To make [the assumption] testable, you may have to back up and unearth a prior assumption in the sequence ... Once you've chosen a big assumption to test, the next step is to design your first experiment to challenge it. Robert Kegan and Lisa Lahey suggest how to turn personal tacit assumptions into explicit testable hypotheses in everyday life.
• Keuth, Herbert [in German] (2004) [Published in German 2000]. "From falsifiability to testability". The philosophy of Karl Popper (1st English ed.). Cambridge, UK; New York: Cambridge University Press. pp. 48–49. ISBN 9780521548304. OCLC 54503549. Consequently, the universal statements, which are contradicted by the basic statements, are not strictly refutable. Like singular statements and probability statements, they are empirically testable, but their tests do not have certain, definite results, do not result in strict verification or falsification but only in temporary acceptance or rejection.
• Popper, Karl (2002) [Published in German 1935; English translation 1959]. The logic of scientific discovery (Reprint ed.). London; New York: Routledge. pp. 95–120 (95). doi:10.4324/9780203994627. ISBN 9780415278447. OCLC 48533950. Theories may be more, or less, severely testable; that is to say, more, or less, easily falsifiable. The degree of their testability is of significance for the selection of theories. In this chapter, I shall compare the various degrees of testability or falsifiability of theories through comparing the classes of their potential falsifiers. This investigation is quite independent of the question whether or not it is possible to distinguish in an absolute sense between falsifiable and non-falsifiable theories. Indeed one might say of the present chapter that it 'relativizes' the requirement of falsifiability by showing falsifiability to be a matter of degree.
• Sober, Elliott (November 1999). "Testability". Proceedings and Addresses of the American Philosophical Association. 73 (2): 47–76 (47–48). doi:10.2307/3131087. JSTOR 3131087. The idea that some experiments really do test a proposition, while others do not, is not controversial, nor does it deserve to be. ... Testing is to testability as dissolving is to solubility. If we can understand what testing is, we also should be able to understand what testability is.