I disagree with almost all of Wade's list and agree with Kyle
Forinash's criticisms.
One more point: According to my understanding of the term,
a hypothesis may be made without regard to whether it is
true or not. Sometimes a hypothesis is assumed, expected,
or hoped to be true ... but sometimes not. In particular,
at the start of a proof by contradiction, we make a hypothesis
that we expect to be entirely false.
To repeat: offering a hypothetical statement says *nothing*
about the truth, falsity, probability, improbability, etc. of
the statement. For more on this, see http://www.av8n.com/physics/hypothesis.htm
There are other words we can use, such as conjecture or
supposition, when we wish to suggest some marginal or
provisional belief in a statement. Of course at this
point we are dealing in shades of gray, since all
scientific laws are to some extent provisional.
I agree with KF that real-world day-to-day scientific
activity is _not_ well described in terms of hypothesis
testing. It is often the case that scientific results
can _retrospectively_ be analyzed in terms of hypotheses
supported and hypotheses rejected ... but this is usually
not necessary or even helpful at the beginning or middle
of a project.
It just cracks me up when people who have not ever done
science, nor even seen it done, insist that "THE scientific
method" equates to hypothesis testing. I wish non-scientists
would refrain from telling scientists how to do science.