From Gilbert Harnam, "The Inference to the Best Explanation," The Philosophical Review (1965):

"In making this inference one infers, from the fact that a certain hypothesis would explain the evidence, to the truth of that hypothesis."

The General Schema of Abductive Arguments:

1. P

2. The best explanation for the fact that P is that Q.

3. Therefore, Q.

('P' represents the data or the phenomena that need explaining.  'Q' stands for a hypothesis that best explains P.)

Factors that make one explanation better than another:

1. Simplicity - the simpler hypothesis is usually true (Ockham's Razor)

2. Plausibility - the hypothesis that is somehow more "plausible" has more of a chance of being true.

3. No Ad Hocery - ad hoc hypotheses have less of a chance of being true

What does 'ad hoc' mean?  'Ad hoc' is latin phrase that means "for the particular purpose at hand."  A hypothesis is said to be ad hoc when it is added to a theory in a complicated and "made up on the spot way" exactly to meet a certain objection.  For instance, at one point in scientific history, scientists perceived certain perturbations in Uranus' orbit.  Newton's laws of motion could not explain them.  Some theorists stuck a complicated addendum on to Newton's laws to explain this.  This new theory was indeed able to explain these perterbations, but the added hypotheses were rather ad hoc.

And, indeed, they turned out to be false.  The perterbations were being caused by the existence of a then unkown planet: Neptune.  Once the existence of Neptune figured in, no ad hoc hypotheses needed to be added to Newton's theory to explain the phenomena.