# Induction

In Logic, one of two distinct systems of argumentation or reasoning. Induction is the logical method that yields probable conclusions.

Inductive reasoning (or argument) was first distinguished from Deductive reasoning by Aristotle. He defined the difference between the two by the direction of logical inference: deduction moved from universal (or more clearly known) propositions, to particular (less clearly known ones), while he defined induction as the inference from particular claims to universal ones.

In contemporary Logic we define the difference based on the logical status of the conclusion: deductive arguments yield necessary conclusions, while inductive arguments yield probable conclusions.

The success, and therefore the evaluation, of induction is dependent on the evidence supporting the conclusion, not the from or structure of the argument. When thinking about the evidence supporting the conclusion we must think about three distinct questions:

1. Sufficiency - is there enough evidence to make the conclusion more likely than not?
2. Relevance - is the evidence relevant to the conclusion?
3. Clarity - is the evidence unambiguous?

If the evidence in an inductive argument is clear, relevant, and sufficient to make the conclusion more likely than not, we call it a Strong argument.

If an inductive argument is determined to be strong and it has all true premises, it is said to be Cogent. If an inductive argument fails to be strong, it is said to have committed an Informal Fallacy which is some error in sufficiency, relevance, or clarity.

Inductive reasoning is the most common form of argumentation (i.e., rational thought) in the academic world. While Deduction is limited to fields like Mathematics, Logic, Philosophy, and Computer Programing, Induction is used in almost every academic field from the natural sciences to the social sciences. Some of the common forms of inductive reasoning include are:

• Arguments from Analogy
• Arguments from Authority
• Arguments from Probability
• Arguments from Statistical Generalization
• Arguments from Signs