Validity in logic depends on the absolute truth of the statement. This is most often represented in math, where factors are compared in various ways to come to a definite conclusion. For example, the variable "X" equals the sum of "Y" and "Z." Logically then, it is also true that "Y" must equal "X" minus "Z." This is true no matter what definite terms you put in place of "X," "Y" or "Z." Measuring terms such as this also can be used to determine the logical validity of a supposed event. For example, if it is known that all ducks quack and that Donald is a duck, then it is a logically valid assumption that Donald the duck quacks.
The validity of a phenomenon, or an observable event, is often tested using the scientific method. This process repeats an event in various situations to determine which parts of the situation are necessary to the process and helps define what actually happens during an observable event. This process starts with asking a specific question about an observable event. Next, research the event and make a prediction about the likely cause of the event in question. Test this hypothesis and analyze the results. If the prediction comes true in full, then it is valid. If the hypothesis is false or only partially true, then a new hypothesis must be constructed and the process must begin again. This method was used to dispel the notion of "spontaneous generation," which supposed that life can spring from inanimate objects such as decaying meat or wheat husks.
One area where validity is extremely relevant is testing. Tests are given to students, potential employees and even the supposed mentally ill to determine the validity of their claim of sanity, employability or graduation. Tests must first be valid before they can be used as a measure of validity. For example, a test of math skill given in English to a Spanish-speaking student is more a validation of the student's inability to speak English than an invalidation of the student's math skills. Tests should be constructed to focus on the specific ability or knowledge the test is trying to validate and not be confused by adding unobserved elements--such as foreign language, obscure references or unrelated subjective questions--to achieve valid results.