¶ deductive reasoning

In contrast to inductive research, deductive research seeks to test a theory or hypothesis based on the observation of data. It begins with the general theory, and gathers data in order to validate or invalidate it. It necessarily leads to a concrete conclusion: the proposition is either true or it is false. Deductive research cannot increase human knowledge; it can only determine if existing knowledge is accurate. It can't be used to predict future events, or to provide insight into unobserved phenomena.

A deductive study begins with a falsifiable hypothesis that will be proven or disproven. In contrast, an inductive study explores open-ended questions to determine what is important and what counts. Bringing these things together is very complicated.

Roger Martin has argued that Western education favours deductive (and inductive) reasoning over abductive reasoning; he suggests that this approach, which accepts a given framework and then tries to apply it to a problem, limits our ability to conceive of novel approaches to a situation. Deductive reasoning, Martin says, focuses on what we think "should be," limiting our view of what is but does not fit the existing model.